Br Heart J 1994;72 (Supplement):S 63-67

Physiological rationale for co-treatment withdiuretics in heart failure

Jay N Cohn

Contemporary treatment of heart failureshould be based on an understanding of itspathophysiology. Only in recent years has thecomplex involvement of many systems in thissyndrome become widely recognised. Thesystemic response to impaired left ventricularfunction generates a wide array of symptomsand physiological derangements that maycontribute to the natural history of the syn-drome. The challenge to clinical investigators,the pharmaceutical industry, and the treatingphysician is to identify and properly use atherapeutic regimen that can successfullycounter the unwanted maladaptive responsesto the cardiac disorder without interferingwith the necessary adaptive responses. In thisreview I examine the rationale for poly-pharmacy in this syndrome.

CardiovascularDivision, DepartmentofMedicine,University ofMinesotaMedical School,Minneapo}is,Minnesota, USAJay N CohnCorrespondence to:Dr Jay N Cohn,Cardiovascular Division,Departnent of Medicine,University of MinnesotaMedical School, Box 488UMHC, 420 DelawareStreet SE, Minneapolis,Minnesota 55455, USA.

Clinical syndromeThe clinical syndrome of heart failure has itsorigin in left ventricular dysfunction. Thisdysfunction may manifest itself predomi-nantly in systole as a reduced wall motion andlowered ejection fraction, or in diastole as

impaired filling characteristics of the left ven-tricle. The end result of systolic dysfunctionusually is progressive dilatation and hypertro-phy of the left ventricular chamber. The con-

sequence of diastolic dysfunction is an

increased end-diastolic pressure. The haemo-dynamic manifestations of these disorders are

similar. Cardiac output may be normal or

only modestly decreased at rest but theincrease in cardiac output normally generatedby exercise is attenuated and exercise isaccompanied by an inordinate rise in left ven-tricular filling pressure. These cardiac mani-festations of heart failure are hallmarks of thedisease process. None the less, it is importantto recognise that considerable left ventriculardysfunction may exist in the absence of symp-toms of heart failure such as impaired exerciseperformance or quality of life.' Thus thehaemodynamic derangement of left ventricu-lar dysfunction is not adequate to explain theclinical syndrome.An inevitable accompaniment of the clini-

cal syndrome of heart failure is an impairmentin exercise performance. Clinically this isrecognised by the patient's inability to carryout his or her normal activities without symp-toms of fatigue or breathlessness. The degreeof exercise intolerance experienced by thepatient (the New York Heart Association clas-sification) is often used to classify the severityof the syndrome.2 As this classification is

based on historical information obtained byinterviewing the patient, rather than by anyobjective measure of exercise performance, itis a rather imprecise and insensitive guide tothe degree of functional impairment. Moreobjective data can be obtained by performinglaboratory exercise tests. Most of these testsare designed to impose a progressive increasein load over time until patients reach theirmaximal exercise capacity.3 Measurement ofgas exchange during exercise can confirm thatthe anaerobic threshold has been surpassedand that the symptoms of dyspnoea andfatigue represent true maximal exercise per-formance.4 This documentation of maximalexercise capacity does not necessarily providemuch insight into the limitation of activity innormal life, which usually consists of submax-imal exercise performed intermittentlythroughout the day. Attempts to develop sub-maximal exercise tests with suitable endpoints to evaluate this impaired lifestylerelated to reduced exercise capacity have notbeen very successful.5A third important manifestation of heart

failure is sodium retention. Oedema and pul-monary congestion are prominent manifesta-tions that usually alert the physician to thepresence of heart failure. Indeed, the diagno-sis is generally not made in the absence offluid retention. Although sodium retention isa common manifestation of heart failure andin some instances may occur as an early signof the disease, the syndrome of left ventriculardysfunction and impaired exercise capacitymay exist without abnormal sodium retention.It is these non-oedematous patients with heartfailure who represent the most troublesomediagnostic challenge to physicians confrontedwith non-specific complaints of fatigue. Whenoedema is present the physician appropriatelygives a diuretic agent that usually relieves thecongested state. In the past, however, manyphysicians assumed that diuresis had effec-tively treated heart failure and did not recog-nise that the progressive syndrome continuedto exist in the absence of fluid retention.

Ventricular arrhythmias are also a commonmanifestation of the syndrome of heart fail-ure.6 The mechanism of these ventriculararrhythmias is not completely understood butthey seem to be directly related to the severityof cardiac dysfunction.7 Most of these ventric-ular arrhythmias in such patients are relativelyasymptomatic, often presenting as unexpectedpremature ventricular contractions on an elec-trocardiogram or as episodes of non-sustainedventricular tachycardia noted on Holter

S 63 on 19 A

ugust 2018 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.72.2_S

uppl.S63 on 1 A

ugust 1994. Dow

nloaded from

Cohn

monitoring. Although the significance of theseventricular arrhythmias is controversial, thehigh incidence of sudden death in thesepatients raises the possibility that the electricalinstability of the heart is playing an importantpart in the natural history of the disease.A fifth manifestation of heart failure is

activation of neurohormonal mechanisms.Plasma noradrenaline (norepinephrine) isalmost invariably raised in patients with heartfailure,8 and is also increased in patients withleft ventricular dysfunction even in theabsence of the symptoms of heart failure.9Plasma renin activity may be strikinglyincreased in patients with heart failure,8 andthere are also increases in plasma argininevasopressin'0 and in plasma atrial natriureticfactor."' The mechanism of activation of theseneurohormonal mechanisms is not entirelyunderstood, although haemodynamic abnor-malities may play a key part. Because of thevasoconstrictor effect of the sympathetic ner-vous system, the renin angiotensin system,and the vasopressin system, these increasedhormonal concentrations are usually thoughtto be related to the vasoconstricted state ofheart failure. Atrial natriuretic factor func-tions in the circulation at least in part as avasodilator and is often viewed as a counterregulatory vasodilator mechanism in the syn-drome of heart failure.'2 As vasoconstriction isa hallmark of the syndrome, it is clear that thevasodilator effect of atrial peptide is inade-quate to counteract the forces of vasoconstric-tion that exist in heart failure.

Determinants of survivalThe severity of heart failure might best becharacterised by the likelihood of dying.Recent trials have provided a data base fromwhich one can assess the relation of thephysiological variables already described tomortality. In most studies the function of theleft ventricle, as assessed by the measurementof left ventricular ejection fraction, has servedas the most powerful predictor of survival.This relation has been best delineated in theVeterans Administration heart failure trial (V-HeFI) in which a close correlation betweenthe depression of left ventricular ejection frac-tion and the annual mortality has been noted.7These data certainly suggest that the left ven-tricle is the culprit organ in heart failure andthat preservation of left ventricular functionmight have a long-term salutary effect on thecourse of the syndrome."3

Although exercise tolerance correlatespoorly with left ventricular ejection fraction,peak exercise capacity measured by gasexchange during bicycle exercise in the V-HeFT trial was directly related to survival inheart failure.7 Thus for any given level of leftventricular ejection fraction poor exercisecapacity was associated with a poorer progno-sis than good exercise capacity. These find-ings raise the possibility that the left ventricle isnot the sole target organ in heart failure andthat other regulatory mechanisms are alsocritical to the natural history of the disease.

The presence of ventricular arrythmias isassociated with a poor prognosis,'4 but theindependent influence of the rhythm distur-bance on mortality can only be evaluated bymultivariate analysis taking into considerationthe other univariate predictors of mortality.With this approach in the second trial (V-HeFT II), the presence of couplets or non-sustained ventricular tachycardia on a Holtermonitor, found in 60% of the patients, was asignificant independent prognostic marker.7

Plasma noradrenaline concentration also isan independent predictor of survival. In previ-ous studies from our laboratory a plasmanoradrenaline concentration >600 pg/mlcarried with it a significantly poorer prognosisthan a plasma noradrenaline concentration<600 pg/ml.'5 These findings have now beenconfirmed in the much larger V-HeFT IIdatabase. In that trial a plasma noradrenalineconcentration <600 pg/ml was associated witha good prognosis, a plasma noradrenalineconcentration from 600 to 900 pg/ml with anintermediate prognosis, and a plasma nora-drenaline concentration of >900 with a poorprognosis.'6 Whether this adverse long-termeffect of heightened plasma noradrenalinemerely identifies a patient population withmore severe haemodynamic derangement orwhether the activated sympathetic nervoussystem is an independent risk factor for pre-mature death remains to be determined.None the less, the data linking the sympa-thetic nervous system to mortality in patientswith heart failure has led to growing enthusi-asm for the use of agents that inhibit theaction of the sympathetic nervous system. fiBlockers have been advocated in an attemptto alter the prognosis of the disease.'7

Assessing the mortality in patients withheart failure represents one way to explore theprogressive nature of the heart failure syn-drome. Diseases -that damage the ventricularmyocardium do not necessarily produce pro-gressive damage. Recently it has become clearthat even in the absence of more directmyocardial damage the process of cardiacdilatation and hypertrophy may proceed. Thisprogressive change in the left ventricle in theabsence of further primary damage may bestbe viewed as a process of ventricular remodel-ling.'8 It is this remodelling process that leadsto dilatation of the chamber and progressivehypertrophy that may be an important riskfactor in the development of pump failure,ventricular arrhythmias, and prematuredeath.'9 The peripheral vasculature also playsa part in this progressive process by imposingan increased impedance on left ventricularejection.20 The vascular impedance in heartfailure may initially be high because of height-ened vasoconstrictor tone in the vasculature,but eventually some of the increased imped-ance may relate to structural changes in thevessel wall that may also be thought to repre-sent a hypertrophy and remodelling process.'9Thus the myocardium and the peripheralvasculature may be viewed as the main targetorgans in heart failure contributing to pro-gression of the syndrome. The role of

S 64 on 19 A

ugust 2018 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.72.2_S

uppl.S63 on 1 A

ugust 1994. Dow

nloaded from

Physiological rationale for co-treatment with diuretics in heartfailure

endocrine factors, particularly local tissueendocrine mechanisms,2' cannot be dis-counted as contributing to both the myo-cardial and peripheral vascular changes overtime.

Regulatory and counter regulatoryprocessesThe complex physiological responses to leftventricular dysfunction may be viewed as reg-ulatory or counter regulatory, depending onthe clinical setting and the degree of theresponse. Thus vasconstriction is compen-satory because it supports arterial pressure inthe face of a falling cardiac output; but itbecomes counter regulatory when it results inan increased impedance that further depressesperformance of the left ventricle. Left ventric-ular dilatation may be regulatory if it allowsmaintenance of a normal stroke volume withreduced wall motion, but it becomes counterregulatory if the dilatation leads to increasedwall stress, inadequate myocardial perfusion,and metabolic or structural abnormalities thatlead to progression of the pump failure.The other physiological abnormalities

found in heart failure may also be viewed asdouble edged swords capable of both support-ing circulatory function and leading to pro-gression of the circulatory deficiency.Hormonal stimulation, sodium retention, leftventricular hypertrophy, tachycardia, etc, canall be viewed as compensatory in some situa-tions and deleterious in others.

Therapeutic implicationsThe multiple physiological mechanisms iden-tified in heart failure encourage the use ofcombinations of agents that can inhibit oralter these mechanisms to influence thecourse of the disease. Diuretics are obviouscandidates for treatment because of the com-mon sodium retention in the syndrome.Indeed, most other drugs used to treat thesyndrome may improve haemodynamics butusually do not eliminate the need for con-comitant diuretic treatment. The loop diuret-ics, such as frusemide, are the most popularagents for use in heart failure and are oftenvery effective in relieving the congestion andoedema. The problem with diuretic treatmentin patients with heart failure is that it usuallydoes not lead to an increase in cardiac outputand may further activate neurohormonalmechanisms that can aggravate vasoconstric-tion in the syndrome. As loss of potassium is acommon consequence of loop diuretic treat-ment, to give supplementary potassium or apotassium retaining diuretic is a prudentapproach to management. Patients with heartfailure often need very high doses of diureticfor adequate diuresis and it is in these patientsthat giving a second diuretic, such as metola-zone, can be most effective. Studies that haveexplored the long-term response of patientstreated with diuretic alone have confirmedthat to give digoxin or angiotensin convertingenzyme (ACE) inhibitor with a diuretic

reduces morbidity.22Vasodilators have become popular in the

management of heart failure because of thehaemodynamic evidence of vasoconstrictionand the finding that acute doses of vasodila-tors can profoundly improve left ventricularfunction.2' This favourable haemodynamicresponse to vasodilators has been the stimulusto develop oral regimens that could be givenchronically for long-term benefit.24 The viewthat the acute haemodynamic response to avasodilator can be translated into long-termefficacy may be naive. Improvement of exer-cise capacity is not temporally related to theimprovement in haemodynamics induced by avasodilator drug.25 Furthermore, some drugsthat exert a potent vasodilator effect have notnecessarily been associated with improvedsymptoms or exercise capacity.26 Thus themechanism by which exercise capacity isimproved seems to be independent of thehaemodynamic response to the vasodilatordrug in heart failure. This improvement ofexercise tolerance may relate more to directimprovement in perfusion of skeletal muscle,to alterations in the structure or metabolicfunction of skeletal muscle, or to a subtlereconditioning process that may take monthsto achieve. None the less, several vasodilatorregimens, including ACE inhibitors,27 oralnitrates,28 and the combination of hydralazineand isosorbide dinitrate,29 have been shown toimprove peak oxygen consumption and maxi-mal exercise capacity in patients with heartfailure.The vasodilator drugs have a clear place in

the contemporary management of patientswith heart failure. Newer vasodilator drugsare being developed in the hope that theymight be effective in combination with otherdrugs in the management of the syndrome.The ACE inhibitors, although they exert avasodilator effect, also have other metabolicand haemodynamic actions that may notresult specifically from their vasodilator mech-anism. Hydralazine and isosorbide dinitraterepresent a potent vasodilator combinationthat is effective both in prolonging life andimproving left ventricular function and exer-cise tolerance. As this generic drug combina-tion is not manufactured and neither of thesedrugs is approved for use in heart failure, theyhave not been widely used in the treatment ofheart failure because of the absence of a mar-keting effort. Consequently, other newervasodilator drugs are currently being devel-oped in the hope that a marketable drug couldbe used in the syndrome. Calcium antagoniststhat have been shown to be more vasoselectivethan previously marketed calcium antagonistsare currently being studied in patients withheart failure. Felodipine and amlodipine arethe most prominent examples of this newclass of calcium antagonists. It is likely thatmost studies with these new agents wouldinvolve patients already receiving ACEinhibitors, in which case the calcium antago-nist would serve as a combination vasodilatorin association with an ACE inhibitor. Anothernew vasodilator compound, flosequinan, has

S 65 on 19 A

ugust 2018 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.72.2_S

uppl.S63 on 1 A

ugust 1994. Dow

nloaded from

Cohn

been shown to improve exercise performanceand quality of life in patients treated with anACE inhibitor, but the dose used in those trialshas shown an adverse effect on survival.30An orally effective positive inotropic agent

has long been sought by the pharmaceuticalindustry to replace or supplement digoxin inthe management of heart failure. The failingleft ventricle shows reduced ventricular wallmotion, and agents with potent inotropiceffects, such as dobutamine and phosphodi-esterase inhibitors, can improve cardiac func-tion in part by increasing contractility of theleft ventricle. Despite the intuitive attractive-ness of an agent that could improve the con-tractile behaviour of a depressed left ventricle,clinical trials with new orally effectiveinotropic agents, mostly phosphodiesteraseinhibitors, have been disappointing.3' Despitethe evidence that these drugs can strikinglyaugment cardiac performance when givenacutely, long-term administration even withsustained heamodynamic efficacy, has notnecessarily been associated with relief ofsymptoms or improved quality of life.Furthermore, one recent study has shown anadverse effect on survival of milrinone givento patients with severe congestive heartfailure.32The failure of these early trials with positive

inotropic drugs may relate more to the toxicside effects of these agents than to any flaw inthe conceptual design of the trials. The searchfor an inotropic agent that is safer, perhaps byvirtue of a different mechanism of action, is ahigh priority in the drug development indus-try. One such drug, pimobendan, seems toexert at least part of its effect by sensitising thecontractile apparatus to calcium. In early tri-als this drug has improved haemodynamicsand exercise tolerance as well as quality of lifein patients with moderate to severe heart fail-ure.33 Additional studies with this drug arenecessary not only to confirm its efficacy butalso to show if it is safe to give long term.The multiple mechanisms involved in the

development and progression of heart failuremake polypharmacy a rational approach to

management. The figure shows that the sitesof action of the drugs currently used in thetreatment of heart failure should be com-plementary in contributing to a favourableresponse to treatment.

Methods of assessmentOne of the main problems of evaluating treat-ment for heart failure is the identification ofthe appropriate end point. Our growinginsight into the syndrome has made it clearthat not all physiological end points or clinicalend points are equally deranged in patientswith heart failure. Consequently, it would beimprudent to use a single end point as a guideto beneficial treatment. In many studies exer-cise tolerance has served as the main endpoint of treatment, but as this measurement isconfined to peak exercise capacity and hasproved remarkably insensitive to treatment itmay not be its most useful marker. Quality oflife may serve as a more appropriate endpoint, but this has needed a questionnaire thatis aimed specifically at patients with heart fail-ure. Such a questionnaire has been developedat the University of Minnesota and currentlyis in use in a number of centres to evaluateresponse to treatments.34 Preliminary trialssuggest that such a patient based question-naire can serve a useful purpose in establish-ing the efficacy of various treatments.3 Leftventricular function also may be a useful mea-surement, particularly because the prognosisof the disease seems to be so closely linked tothe severity of left ventricular dysfunction.Furthermore, a sequential improvement inejection fraction may be a marker for afavourable effect on long-term mortality."3Ventricular arrhythmias could also serve as anend point, as the high incidence of non-sustained ventricular tachycardia makes thesearch for a drug that can reduce thesearrhythmias attractive. The recent cardiacarrhythmia suppression trial (CAST) hasmade it clear that drugs that directly suppressthe arrhythmia are not necessarily safe.'5

Mortality has emerged as the most persua-

LV dysfunctionSome of the diversephysiological mechanismscontributing to heartfailure along with possiblesites of action ofpharmacological agentsused in treatment. LV, leftventricular; EF, ejectionfraction; RBF, renal bloodflow

? ACE inhibitors,? Growth inhibitors

t Impedance

Vasoconstriction

S 66

.4

on 19 August 2018 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.72.2_Suppl.S

63 on 1 August 1994. D

ownloaded from

Physiological rationale for co-treatment with diuretics in heartfailure

sive end point for response to treatment inpatients with heart failure because it is so eas-ily measured and so non-controversial. Thevasodilator combination of hydralazine andisosorbide dinitrate26 and the ACE inhibitorenalapril29 36 have been shown to prolong sur-vival in patients with heart failure. Despite thefavourable effect of these drugs on the courseof the disease, progression of this syndromewas not halted by these agents. At the end offour years in the V-HeFT II trial more than40% of patients had died even while receivinglong-term enalapril treatment.27 Thus the nat-ural history of this disease seems to be pro-gressive and the present treatments have beenineffective in halting this progression. Thispoor long-term experience provides a windowof opportunity for the development of newdrugs that might more effectively inhibit thecounter regulatory processes that lead to pro-gression. Of particular promise in this areawould be agents that would block growth andremodelling of the heart and the vasculature.Safe agents that could improve contractileforce of the myocardium slightly might also beattractive in altering the progression of thedisease. The need for new agents does notnecessarily mean that they will replace olderagents. Diuretics, digoxin, vasodilators, andACE inhibitors will probably continue toserve as the cornerstones for treatmnent ofheart failure. ,B Adrenoceptor blockers areunder intense evaluation for long-term effi-cacy in heart failure. Addition of new drugs tofurther inhibit the progressive process seemsto be the likely strategy. Therefore, thepolypharmacy that already exists today in thissyndrome is likely to become more complex inthe future as we gain further insight into thephysiological mechanisms and pharma-cological responses of these mechanisms tointervention.

1 The SOLVD Investigators. Studies of left ventricular dys-function (SOLVD)-rationale, design, and methods.two trials that evaluate the effect of enalapril in patientswith reduced ejection fraction. Am J7 Cardiol 1990;66:315-22.

2 The Criteria Committee for the New York HeartAssociation. Diseases of the heart and blood vessels, 6th ed.New York: NYHA, 1964.

3 Patterson JA, Naughton J, Pietras RJ, Gunnar RN.Treadmill exercise in assessment of patients with cardiacdisease. Am.7 Cardiol 1972;30:757-62.

4 Cohn JN. Quantitative exercise testing for the cardiacpatient: the value of monitoring gas exchange.Circulation 1987;76(suppl VI):VI1-2.

5 Cohn JN, Rajfer SF. Evaluation of functional capacity inheart failure: a consensus conference. Heart Failure1990;6:169-73.

6 Francis GS. Development of arrhythmias in the patientwith congestive heart failure: pathophysiology, preva-lence, and prognosis. Am Y Cardiol 1986;57:3B-7B.

7 Cohn JN, Johnson GR, Shabetai R, Loeb H, Tristani F,Rector T, et al, for the V-HeFT Veterans AdministrationCooperative Studies Group. Ejection fraction, peakexercise oxygen consumption, cardiothoracic ratio, ven-tricular arrhythmias and plasma norepinephrine asdeterminants of prognosis in heart failure. Circulation1 993;87:VI-5-1 6.

8 Levine TB, Francis GS, Goldsmith SR, Simon A, CohnJN. Activity of the sympathetic nervous system andrenin-angiotensin system assessed by plasma hormonelevels and their relationship to hemodynamic abnormali-ties in congestive heart failure. Am .7 Cardiol 1982;49:1659-66.

9 Francis GS, Benedict C, Johnston DE, Kirlin PC, NicklasJ, Liange C-S, et al. Comparison of neuroendocrine acti-vation in patients with left ventricular dysfunction withand without congestive heart failure. Circulation 1990;82:1724-9.

10 Goldsmith SR, Francis GS, Cowley AW, Levine TB,Cohn JN. Increased plasma arginine vasopressin inpatients with congestive heart failure. J Am Coll Cardiol1983;1: 1385-90.

11 Cody RJ, Atlas SA, Laragh JH, Kubo SH, Covit AB,Ryman KS, et al. Atrial natriuretic factor in normal sub-jects and heart failure patients: plasma levels and renalhormonal and hemodynamic responses to peptide infu-sion. 7Clin Invest 1986;78:1362-74.

12 Carson P, Carlyle PF, Rector TS, Cohn JN.Cardiovascular and neurohormonal effects of atrialnatriuretic peptide in conscious dogs with and withoutchronic left ventricular dysfunction. .7 CardiovascPharmacol 1990;16:305-1 1.

13 Cintron G, Johnson G, Francis G, Cobb F, Cohn JN, forthe V-HeFT VA Cooperative Studies Group. Prognosticsignificance of serial changes in left ventricular ejectionfraction in patients with congestive heart failure.Circulation 1993;87:VI-1 7-23.

14 Fletcher RD, Cintron GB, Johnson G, OmdorffJ, CarsonP, Cohn JN, for the V-HeFT Veterans AdministrationCooperative Studies Group. Enalapril decreases preva-lence of ventricular tachycardia in patients with chroniccongestive heart failure. Circulation 1993;87:Vi-49-55.

15 Rector TS, Olivari MT, Levine TB, Francis GS, Cohn JN.Predicting survival for an individual with congestiveheart failure using the plasma norepinephrine concentra-tion. Am HeartJ 1987;114:148-52.

17 Waagstein F, Hjalmarson A, Varnauskas E, Wallentin I.Effect of chronic beta-adrenergic receptor blockade incongestive cardiomyopathy. Br Heart .7 1975;137:1022-36.

18 Pfeffer MA, Braunwald E. Ventricular remodeling aftermyocardial infarction. Experimental observations andclinical complications. Circulation 1990;81: 1161-72.

19 Francis GS, Cohn JN. Heart failure: mechanisms ofcardiac and vascular dysfunction and the rationale forpharmacologic intervention [review]. FASEB 1990;4:3068-75.

20 Finkelstein SM, Cohn JN, Coffins RV, Carlyle PF, ShelleyW. Vascular hemodynamic impedance in congestiveheart failure. Am J Cardiol 1985;55:423-7.

21 Hirsch AT, Dzau VJ. The tissue renin-angiotensin system:potential role in the pathophysiology of heart failure. Am_7 Cardiol 1990;66:22D-32D.

22 Captopril-Digoxin Multicenter Research Group 1988.Comparative effects of therapy with captopril anddigoxin in patients with mild to moderate heart failure.JAMA 1988;259:539-44.

23 Guiha NH, Cohn JN, Mikulic E, Franciosa JA, Limas CJ.Treatment of refractory heart failure with infusion ofnitroprusside. NEnglJMed 1974;291:587-92.

24 Cohn JN, Mathew KJ, Franciosa JA, Snow JA. Chronicvasodilator therapy in the management of cardiogenicshock and intractable left ventricular failure. Ann InternMed 1974;81:777-80.

25 Franciosa JA, Goldsmith SR, Cohn JN. Contrastingimmediate and long-term effects of isosorbide dinitrateon exercise capacity in congestive heart failure. Am.Med 1980;69:559-66.

26 Cohn JN, Archibald DG, Ziesche S, Franciosa JA,Harston WE, Tristani FE et al. Effect of vasodilator ther-apy on mortality in chronic congestive heart failure.Results of a Veterans Administration cooperative study(V-HeFT). N EnglyMed 1986;314: 1547-52.

27 Captopril Multicenter Research Group. A placebo-controlled trial of captopril in refractory chronic conges-tive heart failure. JAm Coll Cardiol 1983;2:755-63.

28 Leier CV, Huss P, Magorien RD, Unverferth DV.Improved exercise capacity and differing arterial andvenous tolerance during chronic isosorbide dinitratetherapy for congestive heart failure. Circulation 1983;67:817-22.

29 Ziesche S, Cobb FR, Cohn JN, Johnson G, Tristani F, forthe V-HeFT Veterans Administration CooperativeStudies Group. Hydralazine and isosorbide dinitratecombination improves exercise tolerance in heart failure:results from V-HeFT I and V-HeFT II. Circulation1993;87:VI-56-64.

30 Massie BM, Berk MR, Brozena S, Elkayam U, Plehn J,Kukin M, et al, for the FACET investigators. Can fur-ther benefit be achieved by adding a vasodilator to tripletherapy in CHF: results of the flosequinan-ACEinhibitor trial (FACET) [abstract]? Circulation 1992;86:I-645.

31 Cohn JN. Inotropic therapy for heart failure: paradisepostponed. N Engl Med 1989;11:729-31.

32 Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ,DiBianco R, Zeldis SM, et al, for the PROMISE StudyResearch Group. Effect of oral milrinone on mortality insevere chronic heart failure. N Engl Jf Med 1991;325:1468-75.

33 Kubo SH, Gollub S, Bourge R, Rahko P, Cobb F, JessupM, et al, for the Pimobendan Multicenter ResearchGroup. Beneficial effects of pimobendan on exercise tol-erance and quality of life in patients with heart failure:results of a multicenter trial. Circulation 1992;83:942-9.

34 Rector TS, Kubo SK, Cohn JN. Patient's self-assessmentof their congestive heart failure. Part 2. Content, reliabil-ity and validity of a new measure, the Minnesota livingwith heart failure questionnaire. Heart Failure 1987;3:198-209.

35 The Cardiac Arrhythmia Suppression Trial (CAST)Investigators. Preliminary report: effect of encainide andflecainide on mortality in a randomized trial of arrhythmiasuppression after myocardial infarction. N Engl J Med1989;321:406-12.

36 The SOLVD Investigators. Effect of enalapril on survivalin patients with reduced left ventricular ejection frac-tions and congestive heart failure. N Engl J Med 1991;325:293-302.

S 67 on 19 A

ugust 2018 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.72.2_S

uppl.S63 on 1 A

ugust 1994. Dow

nloaded from