Immunosuppressive treatment in myocarditis

International Elsevier

Journul of Cardioloa,

22 (1989) 329-338

329

IJC 00818

Immunosuppressive

treatment

A. Salvi, A. Di Lenarda,

L. Dreas, F. Silvestri

Department

of Cardlologv und of ’ Patholoy,,. (Received

’ and F. Camerini

Ospedule Mcrggiore and University of Trieste, Trleste, Itu(l

15 June 1988; revision

Salvi A, Di Lenarda A, Dreas L, Silvestri J Cardiol 1989;22:3299338.

in myocarditis

F, Camerini

accepted

22 August

1988)

F. Immunosuppressive

treatment

in myocarditis.

Int

Twenty patients (13 males and seven females) with a biopsy-proven diagnosis of myocarditis underwent a The primary objective of the study was the period of treatment with prednisone and azathioprine. observation of histologic changes which occur during treatment and after treatment withdrawal. The secondary objective was endomyocardial biopsies

the detection, were obtained

if any, of changes and

the

treatment

in left ventricular ejection fraction. was adjusted in order to achieve

Multiple complete

disappearance of the myocardial inflammation. The histologic status was improved in all patients, although complete disappearance of the signs of active disease was seen in 15 patients only. Two patients died during the observation lished in some

period. patients

A clear relationship between histologic (50% of all cases showed a worsening

status and immunosuppression was estabafter withdrawal from the treatment). An

overall improvement of the ejection fraction was observed (from 0.37 k 0.14 to 0.46 k 0.17), but a direct effect of the treatment on the recovery of ventricular function cannot be stated. In some patients, however, a direct relationship between the histological changes and the changes in ejection fraction was seen. These data suggest biopsy-proven Key words:

that treatment with prednisone and azathioprine myocarditis and depressed ventricular function.

Endomyocardial

biopsy;

Myocarditis:

Introduction Myocarditis is clinically suspected when congestive heart failure, atrioventricular block, or ventricular arrhythmias develop rapidly in a previously normal subject, particularly, after a viral illness [l-3]. Various non-invasive diagnostic tests may help in the clinical diagnosis [4-71 but only since the

Correspondence to: A. S&i. Department of Cardiology, Ospedale Maggiore. Piazza Ospedale 1, 34129 Trieste, Italy. Presented as a Communication at the international Symposium “New Concepts in Viral Heart Disease”. Munich, 25-28 May. 1988. 0167.5273/89/$03.50

li’ 1989 Elsewer

Science Publishers

may

be beneficial

in some

patients

with

Immunosuppression

technique of endomyocardial biopsy has become available [8], it is possible to obtain histologic confirmation of the disease in vivo. The similarity between the histologic picture in acute myocarditis and that seen in acute cardiac rejection suggested that immunosuppression might benefit both conditions. and some authors have already reported their observations [9-131. A beneficial effect of immunosuppression in acute myocarditis is, however, still unproven. The aim of our study was to evaluate the effect of immunosuppressive treatment with prednisone and azathioprine on both histology and left ventricular function in 20 patients with biopsy-proven acute myocarditis.

B.V. (Biomedical

Division)

330

Patients Up to January 1988. endomyocardial biopsy was performed in 420 patients admitted to our institution for the investigation of a suspected myocardial disease. Myocarditis was diagnosed histologically in 38 patients. Twenty patients in whom the diagnosis was made within the last four years. and who entered a protocol on the effects of immunosuppressive treatment in myocarditis of unknown etiology, are the basis of this study. This group consists of 13 males and 7 females with an age range of 16-66 years (mean 33.6). Clinical picture The clinical presentation was congestive heart failure in 16 patients, persistent atria1 tachycardia and ventricular premature complexes in two and typical angina pectoris at rest in two others. A history compatible with a viral illness in the preceding four months was found in nine patients. In four patients the illness was diagnosed within two months after delivery. Seventeen patients had the histologic diagnosis made within five months of clinical presentation (mean 2.2 months, range O-5). Seventeen patients showed a depressed (less than 0.50) left ventricular ejection fraction at diagnosis. Two patients with a history of angina and one with a history of atria1 tachycardia had normal or minimally depressed ejection fraction (normal values for ejection fraction in our laboratory are 0.67 f 0.09). Selective coronary arteriography. in order to exclude coronary artery disease was performed in all patients over 25 years of age, except in two women aged 34 years who presented with congestive heart failure. Both coronary arteries were opacified during aortography in the three youngest patients. Methods Endomyocardial

biopsy

Four or five samples of tissue from the right or the left ventricle

were obtained with a King’s

endomyocardial bioptome [14] or the Cordis adaptation of this instrument [15] and immediately fixed in 10% buffered formalin. Multiple 5 pm thick sections were made, stained with haematoxylin eosin. Masson’s trichrome and Weigertt Van Gieson and examined by light microscopy. Histologic

classification

The methods suggested by a panel of pathologists who met at Dallas in 1984 [16] were used for the diagnosis of myocarditis. All cases were retrospectively examined and the diagnosis was accepted when two different observers agreed. Subsequent biopsies were classified according to the above mentioned paper. Immunosuppressive

treatment protocol

Immunosuppressive treatment consisted of oral azathioprine, 75 mg per square metre per day for at least six months (unless the white blood cells count fell below 4000/mm7) and oral prednisone 50 mg per square metre per day for two weeks. then tapered until withdrawal of the drug after at least six months of treatment. In patients admitted to the study after June 1985, the prednisone dose of 0.3 mg per kilogram per day. when reached, was maintained for approximately four months. In the event of persistent active myocarditis during treatment or reactivation after interruption, the prednisone dosage was increased either to the initial dose or to the one previously demonstrated effective. In all patients. the interval between the onset of symptoms and the beginning of treatment was at least one month. The usual means of treatment of congestive heart failure. arrhythmias, diabetes and hypertension were continued in addition to the immunosuppression. Treatment with digoxin. diuretics. vasodilators and amiodarone, when already present. was continued. In seven patients, metoprolol 50-700 mg per day was added during the period of observation for either persistent sinus tachycardia at rest (five cases) or hypertension (two cases).

331

Patients’ assessment

Results

The study was originally designed to be based on serial endomyocardial biopsies performed after approximately two months of continued treatment and at least two weeks after withdrawal of the immunosuppressive treatment. Multiple endomyocardial biopsies were, however, carried out in several patients who showed a persistent myocarditis. In these patients, the periods of treatment and follow-up were consequently longer. Left ventricular ejection fraction was determined by angiography in 15 patients and by gated cardiac blood pool scintigraphy technetium-99 in two patients. Data used for comparison during follow-up were always obtained with the same technique. Statistical

analysis

Where appropriate, results was performed

statistical analysis of the using Wilcoxon’s test.

Endomyocardial

biopsy

A mean of five endomyocardial biopsy procedures (range 3312) was performed in each patient during the study and an average number of 4.2 (range 3-7) samples was obtained. In three patients, however, one of the biopsies obtained during follow-up consisted of only three samples useful for histologic evaluation. In all patients (Fig. l), a control endomyocardial biopsy (second biopsy) was performed after approximately two months of treatment (mean 73.5 days, range 45-150 days) while the mean prednisone dose was about 50 mg per day. Although some improvement was apparent in the majority of patients, a persistent myocarditis was present in seven patients and they were treated again with a higher dosage of prednisone. The treatment was continued, but fapered, and about five months later (last biopsy), at a mean

myocardl t Is (1st EMB) or persistent myocardltls (other EMB) A

healing myocardltls healed myocardltls

1st

biopsy

Fig. 1. Histologic

changes

2nd

during

biopsy

and after immunosuppressive

last

treatment.

biopsy

IXI= period

therapy dlSConflnued

of immunosuppressive

treatment.

332

lzF% 45: 4035302520-

Fig. 2. Changes

of histology

and ejection

1St

Fig. 3. Histologic

fraction

and relationship with the immunohupprsssive as in Fig. 1.

treatment

dlscontinuatlon of therapy

2nd dlscontlnuatlon of therapy

3 rd dlscontlnuatlon Of therapy

mean

maan

mean

changes

300 days

502

days

in one patient.

*. A. and

l

795 days

during and after Immunosuppressive treatment in seven patients who needed reinstitution first discontinuation (one died before being rebiopsied). *, A. 0 and IXIas in Fig. 1.

of therapy

after

333

prednisone dose of 25 mg per day. the results were definitely better and all patients examined showed a clear reduction of the histologic signs of activity. It has to be noted. however. that data for comparison were available in 15 patients only. Four patients with healing or healed myocarditis after two months of therapy reduced the drug dosages as scheduled but refused a subsequent biopsy during maintainance. In one patient, the treatment was stopped after four months when he presented with appendicitis complicated by peritonitis. In 16 patients. the histologic status was reassessed two weeks after withdrawal of treatment and about nine months (mean 285 days, range 120-510 days) after entry into the study (Fig. 1). One patient died before being restudied and three refused further invasive investigations. A worsening of histology was seen in about 50% of the patients who had improved during treatment and two showed a severe reactivation. All were treated at doses previously demonstrated effective or even

higher, except one who died shortly after. The treatment, very prolonged in some patients (Fig. 2) was eventually associated with histology of healing or healed myocarditis (Fig. 3). Left ventricular

ejection fraction

Ejection fraction values determined before treatment were compared to values obtained in the latest period of follow-up in 17 patients, either during treatment (in three patients) or after treatment withdrawal (in 14 patients) (Fig. 4). Two patients refused further investigation while ejection fraction at entry could not be estimated in one case. An overall increase (from 0.37 k 0.14 to 0.46 + 0.17, P < 0.02) was observed and in individual cases a clear improvement was evident. The changes in ejection fraction were not usually evident in the early months of immunosuppressive treatment and the major improvement

EM 70

!

/

50

40

30

20

p -z 0.02

I

before

end of treatment

treatment

Fig. 4. Changes

of ejection

fraction

after withdrawal

(14 patients)

or during

treatment

(3 patients).

t = patients

who died.

334

60

SO.

40

30.

20.

10 4

n= 17

0

Fig. 5 Changes

of ejection

fraction

3

during

6

prolonged

9

period

12

18

of immunosuppressive died.

was often observed after several months of continuous treatment or sometimes after several unsuccessful attempts of treatment withdrawal (Figs. 2, 5). Immunosuppressive

15

treatment and side effects

All patients became clinically “cushingoid” with the doses of prednisone employed. Two of the oldest patients developed hypertension which required treatment with metoprolol and captopril and in one of them insulin was given to control a previously unknown diabetes mellitus. In the same case, suboptimal doses of azathioprine were used due to recurrent leukopenia. The same patient developed, after two years of treatment, oral candidiasis and epididymitis, which were controlled with short courses of topical mycostatin and oral tetracyclin. In one patient, the treatment was stopped after four months with the onset of an acute peritonitis as a complication of a previously known chronic appendicitis which perfo-

21

24 27 months

treatment

in seventeen

patients.

i= patients who

rated. Interestingly, he had normal white blood cell counts during the period of treatment which raised, with normal differentials, during the acute infection. He was surgically treated and made an uneventful recovery. Immunosuppression was not restarted because myocarditis was demonstrated to be completely healed by endomyocardial biopsy one year later. Mortality Two patients died. Both had low ejection fraction which deteriorated slowly during the observation period (in both from 0.26 to 0.16). Another patient died after the end of treatment for noncardiac causes (cerebral haemorrhage during anticoagulant treatment). Discussion Myocarditis acterized by

an

is defined “as inflammatory

a process charinfiltrate of the

335

myocardium with necrosis and/or degeneration of adjacent myocytes not typical of the ischemic damage associated with coronary artery disease” P61. Although different organisms (bacteria, viruses, rickettsia, fungi, parasites) may cause the disease. in most cases a viral etiology is generally suspected [17,18]. The clinical diagnosis of myocarditis is often difficult and sometimes impossible. The disease may be suspected when congestive heart failure, ventricular arrhythmias or atrioventricular block [l-3,19.20] develop rapidly in a previously normal subject. Another. although rare, presentation is that of acute myocardial infarction in a young patient, particularly, after a viral illness [21]. It has been claimed that different noninvasive diagnostic tests [4,7] may help or even confirm [5,6] the clinical suspect. However. it is possible to obtain histologic confirmation of the disease in vivo, only since the technique of endomyocardial biopsy has become available [S]. Efforts aimed at the identification of viruses in the myocardium of patients with myocarditis almost invariably failed [22,23]. This is at variance

Fig. 6. Endomyocardial

biopsy

(haematoxylin

with the evidence for a viral implication in up to 40% of cases of myocarditis and pericarditis [17] and with the finding of rising serum titres for coxsackie B virus neutralizing antibodies in up to 50% of patients with biopsy proven myocarditis [12]. The concept of an immune-mediated damage, after viral infection, based on the experimental work of Wilson [24] and Woodruff and Woodruff [25] was therefore suggested also in the pathogenesis of human myocarditis. From this concept and from the observation that a similarity exists between the histologic aspect of myocarditis and that of cardiac allograft rejection. in which immunosuppressive drugs are beneficial, comes the rationale for this form of treatment. After the initial report of the Stanford group [9] several other authors published their results in the treatment of myocarditis with immunosuppressives [lo-13,26,27]. Different histologic criteria for the diagnosis of myocarditis were, however, used and great caution should be exercised when interpreting the results. The need for a more uniform diagnostic and classification system was therefore felt.

and eosin, original magnification loss and necrosis of myocytes.

X

100). Diffuse

mononuclear

- cells infiltrate

with

336

Fig. 7. Endomyocardial

biopsy (haematoxylin and eosin, original applied to the sarcolemma with irregular

A Multi Centre Myocarditis Trial, based on uniform and strict histologic criteria, known as the Dallas criteria [16], was therefore designed [28] in order to clarify the role of immunosuppression in myocarditis, not convincingly defined by the initial experiences. Until the results of this trial are available, it may be useful to report our observations. All patients included in this study showed on their initial endomyocardial biopsy “both an inflammatory infiltrate and damage of the adjacent myocytes . . in the form of frank necrosis (Fig. 6) myocyte vacuolization, irregular cellular outlines or cellular disruption with lymphocytes closely applied to the sarcolemma” (Fig. 7) [16]. After a retrospective analysis of cases initially reported [29], two patients were excluded due to the fact that the histologic findings were considered insufficient for a definite diagnosis of active myocarditis. During follow-up, myocarditis was classified as persistent, healing or healed [16]. The study was primarily designed to follow the histologic changes during and after an immunosuppressive treatment with prednisone and azathioprine and to observe

magnification x 100). Diffuse mononuclear cellular outlines and necroslr of myocytes.

- cells infiltrate

closely

if any change in left ventricular ejection fraction takes place during follow-up. Six months was the originally planned duration of treatment, but the histologic response led to adjustments in dosage and duration of immunosuppression. Most patients had a period of followup longer than six months and some were treated for as long as two years. An important limitation is that it cannot be stated whether the observed changes were secondary to the immunosuppression or spontaneous. This problem could be probably solved by the randomized study which is presently in progress [28,30]. The behaviour of individual patients (Fig. 2) and the fact that 50% of patients who healed histologically during therapy showed a worsening after interruption of the treatment (Fig. 1). however, suggests a direct link between immunosuppressive drug regimen and the histologic findings observed. These data suggest also that treatment should not be carried out according to rigid protocols but rather modulated according to histology. After two months of high continuous prednisone therapy, one-third of the cases was still unchanged. It

337

may be noted also that at the second control biopsy, performed during a low prednisone dose after approximately seven months of treatment, none of the patients showed signs of persistent myocarditis. A benefit related to a prolonged period of treatment at low maintainance doses as in other immuno-mediated disease [31] is therefore postulated. Most previous reports on immunosuppression in myocarditis did not provide an objective and reliable evaluation of ventricular function in all patients [9-11,271. In the study of Hosenpud et al. any improvement, ]131, who did not observe ventricular function was assessed by right heart catheterization and M-mode echo data, but Daly et al. [12] provided evidence for an improvement of ventriculographic ejection fraction after treatment. Whether this improvement is secondary to the treatment is, however, still undefined, because Kawamura et al. [32] have shown that 38% of patients with biopsy-proven myocarditis improve spontaneously and because other authors [26] have also noted a spontaneous improvement of ejection fraction in patients in whom myocarditis, although unequivocally diagnosed, was not treated with immunosuppression. An overall significant improvement of ejection fraction was observed in our patients. Interestingly, this was more evident in cases treated for longer periods, because of frequent reactivation of myocarditis (Fig. 5). The behaviour of individual cases (Fig. 2) may suggest a direct effect of the immunosuppression both on histologic healing, and on recovery of ventricular function. This is in accordance with some other observations of Mason et al. [9], Daly et al. 1121 and Dee et al. [26] and suggests that there is a subset of patients who improve as a result of the immunosuppression and show a relapse when the drugs are reduced or discontinued. One limitation of this study is that it was not conceived as a controlled trial, although the need for a randomized study to assess this form of treatment had already been suggested [12]. Another potential limitation is that the autoimmune pathogenesis of human myocarditis has become less convincing after the experimental work of Bowles et al. [33]. These authors [33] were able

to detect coxsackie B virus specific RNA in myocardial tissue of patients with myocarditis even in their late, healed stage and this may imply continuing virus replication in the myocardial cells. Should this observation be confirmed, the rationale for the use of an immunosuppressive treatment in human myocarditis should be redefined. Finally, it has to be mentioned that, although the traditional measures of treatment were maintained constant during the period of study, betablockers were added in seven patients. This form of therapy may have a potential beneficial effect in patients with depressed ventricular function secondary to dilated cardiomyopathy [34,35]. We cannot define whether the complex effects induced by beta-blockade [36-381 benefitted also our group of patients with myocarditis. In conclusion, (1) Immunosuppressive treatment is usually associated with a reduction of histologic signs of active myocarditis. (2) Reduction or interruption of treatment is often associated with a worsening histology (and sometimes with a deterioration of left ventricular function). (3) Immunosuppressive treatment is associated, in a subgroup of patients, with an improvement of left ventricular function. (4) The improvement of left ventricular function may appear late. (5) The relationship between the treatment and the improvement in histology and in ventricular function and the possible deterioration after withdrawal (observed also by other authors) [9,12,26] suggest that immunosuppressive therapy may be useful at least in a subset of patients with active myocarditis.

References Take M. Sekiguchi M, Hiroe M. Hirosawa K. Early clinical profiles of cases with histopathologically proven acute idiopathic myocarditis and a proposal for diagnostic criteria. Jpn Circul J 1981:45:1415-1420. Onouchi Z, Haba S, Kiyosawa N, Shimazu S, Hamaoka K. Kusunoki T. Stokes-Adams attacks due to acute nonspecific myocarditis in childhood. Jpn Heart J 1980:21: 307-315. Kitahura Y. Morita H. Virus myocarditis and cardiomyopathy. Jpn Circul J 1979;43:1017-1031. Hauser AM. Gordon S, Cieszkowski J, Timmis GC. Severe transient left ventricular “ hypertrophy” occurring during acute myocarditis. Chest 1983;83:275-277. O’Connell JB. Henkin RE, Fowles RE. Modes of detection

338

6

I

8

9

10

11

12

13

14 15

16

17 18

19

20

21

22

of inflammation in the cardiomyopathic heart. In: Robinson JA. O’Connell JB. eds. Myocarditis: precursor of cardiomyopathy. Toronto: Collamore Press, 1983;109%127. Yasuda T. Palacios IF. Dee GW, et al. lndium lll-monoclonal antimyosin antibody imaging in the diagnosis of acute myocarditis. Circulation 1987:76:306311. Chandraratna PAN, Nimalasuriya A, Reid CR, Cohn S, Rahimtoola SH. Left ventricular asynergy in acute myocarditis: simulation of acute myocardial infarction. J Am Med Assoc 1983;250:1428-1430. Sakakibara S. Konno S. Endomyocardial biopsy. Jpn Heart J 1962;3:537-543. Mason JW, Billingham ME. Ricci DR. Treatment of acute inflammatory myocarditis assisted by endomyocardial biopsy. Am J Cardiol 1980:45:1037-1044. Zee-Cheng C-S, Tsai CC, Palmer DC, Codd JE. Pennington DG, Williams GA. High incidence of myocarditis by endomyocardial biopsy in patients with idiopathic congestive cardiomyopathy. J Am Co11 Cardiol 1984;3:63-70. Edwards WD. Holmes DR, Reeder GS. Diagnosis of active lymphocytic myocarditis by endomyocardial biopsy: quantitative criteria for light microscopy. Mayo Clin Proc 1982:57:419-425. Daly K. Richardson PJ. Olsen EGJ, et al. Acute myocarditis: role of histological and virological examination in the diagnosis and assessment of immunosuppressive treatment. Br Heart J 1984;51:30-35. Hosenpud JD, McAnulty JH, Niles NR. Lack of objective improvement in ventricular systolic function in patients with myocarditis treated with azathioprine and prednisone. J Am Co11 Cardiol 1985:6:797-801. Richardson PJ. King’s endomyocardial bioptome. Lancet 1974: i:660-661. Richardson PJ. Endomyocardial biopsy: technique and evaluation of a new disposable forceps and catheter sheath system. In: Bolte HD. ed. Viral heart disease. Berlin: Springer, 1984:173-176. Aretz HT, Billingham ME, Edwards WD, et al. Myocarditis: a histopathologic definition and classification. Am J Cardiovasc Path01 1986;1:3-14. Grist NR, Bell EJ. A six year study of Coxsackievirus B infections in heart disease. J Hyg 1974:73:165-172. Richardson PJ. Immunosuppressive therapy in the management of biopsy proven acute myocarditis. Hera 1985;lO: 36-43. Vignola PA, Aonuma K. Swaye PS. et al. Lymphocytic myocarditis presenting as unexplained ventricular arrhythmias: diagnosis with endomyocardial biopsy and response to immunosuppression. J Am Co11 Cardiol 1984;4:812-819. Salvi A, Della Grazia E, Silvestri F, Camerini F. Acute rickettsial myocarditis and advanced atrioventricular block: diagnosis and treatment aided by endomyocardial biopsy. Int J Cardiol 1985;7:405-409. Costanzo-Nordin MR. O’Connell JB. Subramanian R. Robinson JA, Scanlon PJ. Myocarditis confirmed by biopsy presenting as acute myocardial infarction. Br Heart J 1985;53:25529. Lowry PJ. Edwards CW. Nagle RE. Herpes-like virus particles in myocardium of patient progressing to congestive cardiomyopathy. Br Heart J 1982;48:501-503.

23 Engblom E, Ekfors TO, Meurman OH. Toivanen A. Nikoskelainen J. Fatal influenza A myocarditis with isolation of virus from the myocardium. Acta Med Stand 1983;213:75-78. 24 Wilson FM. Miranda QR. Chason JL. Lerner AM. Residual pathologic changes following murine coxsackie A and B myocarditis. Am J Path01 1969:55:253-265. 25 Woodruff JF. Woodruff JJ. Involvement of T lymphocytes in the pathogenesis of coxsackie virus B heart disease. J lmmunol 1974;113:1726-1734. 26 Dee GW. Palacios IF. Fallon JT. et al. Active myocarditis in the spectrum of acute dilated cardiomyopathies: clinical features. histologic correlates. and clinical outcome. N Engl J Med 1985:312:8X5-890. 27 Fenoglio JJ, Ursell PC, Kellogg CF. Drusin RE, Weiss MB. Diagnosis and classification of myocarditis by endomyocardial biopsy. New Engl J Med 1983;308:12-18. 28 Mason JW. Endomyocardial biopsy: the balance of success and failure. Circulation 1985;71:185-188. 29 Salvi A, Hrovatin E, Dreas L, Silvestri F. Camerini F. Changes in histology and left ventricular ejection fraction during immunosuppressive treatment in active myocarditis. Eur Heart J 1987~8 (suppl):267-269. 30 Anderson JL. Fowles RE. Unverferth DV. Mason JW Immunosuppressive therapy of myocardial inflammatory disease. Initial experience and future trials to define indication for therapy. Em Heart J 1987:8(suppl):263-266. 31 Hahn BH. Systemic lupus erythematosus. In: Braunwald E, Isselbacher KJ. Petersdorf RG, Wilson JD, Martin JB. Fauci AS. eds. Harrison’s principles of internal medicine. New York: McGraw-Hill. 1987:1418-1423. 32 Kawamura K. Kitaura Y. Morita H, Deguchi H, Kotaka M. Viral and idiopathic myocarditis in Japan: a questionnaire survey. In: Sekiguchi M, Olsen EGJ, Goodwin JF. eds. Myocarditis and related disorders. Tokyo: SpringerVerlag. 1985:18%22. 33 Bowles NE, Olsen EGJ. Richardson PJ. Archard LC. Detection of Coxsackie-B-virus specific RNA sequences in myocardial biopsy samples from patients with myocarditis and dilated cardiomyopathy. Lancet 1986: i: 1120-l 122. F, Hjalmarson A. Swedberg K. Wallentin I. 34 Waagstein Beta-blockers in dilated cardiomyopathies: they work. Eur Heart J. 1983;4 (suppl. A):173-178. 35 Engelmaier RS, O’Connel JB, Walsh R. Rad N. Scanlon PJ. Gunnar RM. Improvement in symptoms and exercise tolerance by metoprolol in patients with dilated cardiomyopathy: a double blind randomized, placebo controlled trial. Circulation 1985:72:536-546. 36 Fowler MB. Bristow MR. Laser JA, Ginsburg R, Scott LB. Schroeder JS. Beta blocker therapy in severe heart failure: improvement related to beta I-adrenergic receptor up-regulation? Circulation 1984;70: II-1 12. 37 Waagstein F. Hjalmarson A, Varnauskas E. Wallentin I. Effect of chronic beta-adrenergic receptor blockade in con gestive cardiomyopathy. Br Heart J 1975;37:1022-1036. J. Grossman W. Are beta-adrenergic blocking 38 Alderman drugs useful in the treatment of dilated cardiomyopathy? Circulation 1985;71:854-857.