DOI: 10.1111/eci.12483

ORIGINAL ARTICLE Influence of different aetiologies on clinical course and outcome in patients with dilated cardiomyopathy € tz Gelbrich‡,b and Bernhard Maisch§,b € ers†,a, Anette Richter*, Volker Ruppert*, Go Sabine Pankuweit*,a, Claus Lu on behalf of the German Competence Network Heart Failure Department of Internal Medicine – Cardiology, Angiology, Intensive Care and Prevention, University of Marburg, Marburg, Germany, †Department of Cardiology, Klinikum Oldenburg, University of Oldenburg, European Medical School, Oldenburg, € rzburg, Wu € rzburg, Germany, ‡Institute of Clinical Epidemiology and Biometry (ICE-B), Julius Maximilian University of Wu €ßzentrum Marburg (HGZ), Marburg, Germany Germany, §Herz- und Gefa *

ABSTRACT Background The clinical phenotype dilated cardiomyopathy is assumed to be the endstage of a multifactorial aetiopathogenetic pathophysiology which includes a not satisfactorily defined group of patients with inflammatory cardiomyopathy. Methods Within the German Competence Network Heart Failure patients with heart failure due to dilated cardiomyopathy of viral/inflammatory (DCMi/v) and nonviral/noninflammatory (DCM) aetiology were enrolled. After 1 year 237 patients (180 male/57 female) were re-examined including complete clinical work-up. The association of different clinical courses with the time from initial diagnosis of heart failure (newly: ≤ 1 year; late: > 1 year) was investigated. Results After 1-year-follow-up New York Heart Association (NYHA) class (by 0
48 in newly diagnosed DCM and 0
82 in newly diagnosed DCMi/v in addition to 0
24 in late diagnosed DCM and 0
17 in late diagnosed DCMi/v) as well as left ventricular ejection fraction (+14% in newly diagnosed DCM and DCMi/v and +6% in later diagnosed DCM and DCMi/v) were significantly improved in all patients. In patients with early diagnosed dilated cardiomyopathy a strong improvement of NYHA class could be demonstrated. Conclusions This study demonstrates for the first time a significant interaction between duration of disease, NYHA class and left ventricular ejection fraction in patients with DCM. Our results clearly demonstrate that in patients with DCM an early diagnosis within 1 year after occurrence of clinical signs is associated with a strong improvement in the clinical course, whereas late diagnosis results in a loss of change in clinical course and outcome. Keywords Aetiology, dilated cardiomyopathy, epidemiology, heart failure, inflammation, myocarditis. Eur J Clin Invest 2015; 45 (9): 906–917

Introduction The prevalence of dilated cardiomyopathy (DCM) in the general population can only be estimated and clearly varies with age and geography. A minimum of 25% of patients in Western populations have evidence for familial disease with predominantly autosomal dominant inheritance [1–3]. The clinical phenotype of DCM is assumed to be the endstage of a multifactorial aetiopathogenesis and common terminal pathophysiology. Dilated cardiomyopathy comprises poorly defined subgroups of patients with inflammatory-infectious, inflammatory-noninfectious and noninflammatory patients after the exclusion of First

906

a

and last authors

b

contributed equally to this manuscript.

accepted other causes of heart failure such as coronary artery disease, hypertension and valvular diseases [4,5]. Precipitating factors may include predisposition for viral infections, enhanced autoimmunity and environmental factors in addition to a specific ‘genetic background’ of the individual patient [6,7]. DCM can be a late stage of heart failure following infection and inflammation of myocardial interstitial and microvascular cells. In addition to active or fulminant myocarditis, by definition an acute inflammatory disorder of the heart, inflammatory DCM is defined by the chronic presence of inflammatory cells in association with an increase in the left ventricular end-diastolic diameter (LVEDD) or volume (LVEDV), reduced left ventricular ejection fraction (LVEF) and from a more clinical point of view with (New York Heart Association) NYHA class II to IV [8,9].

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation

INFLUENCE OF DIFFERENT AETIOLOGIES

It has been estimated that a family history indicative of DCM (fDCM) is present in 25% to over 50% of patients with dilated cardiomyopathy [10–13]. However, next to a distinct clinical definition, the classification, diagnosis and treatment of DCM associated with chronic inflammation (DCMi/v) or without chronic inflammation (DCM) continue to prompt an eminent debate. The use of endomyocardial biopsy is required to confirm the diagnosis of myocarditis and identifies the underlying aetiology and the type of inflammation (e.g. giant cell, eosinophilic myocarditis, sarcoidosis), which may imply different treatment and prognosis. But up to date, less is known with respect to the influence of the time-course of the initial diagnosis on NYHA class, LVEF and LVEDD. To give an insight into these fundamental interrelations we initialized this project in the framework of the German Competence Network Heart Failure with the hypothesis that an early diagnosis of DCM would influence the clinical course and outcome of DCM.

Methods Study design Between January 2004 and December 2009 322 eligible patients with dilated cardiomyopathy were consecutively enrolled within three subprojects (TP9a n = 272 patients, TP9b n = 11 patients and TP12 n = 39 patients) of the German Competence Network Heart Failure. Coronary artery disease was excluded in all patients after coronary angiography and furthermore, patients with systemic inflammatory or autoimmune disease, sarcoidosis and amyloidosis were excluded. (Fig. 1; Table 1) The protocol was approved by the German Competence Network Heart Failure and conducted in accordance with the principles of the declaration of Helsinki (1996), the International Conference on Harmonization Good Clinical Practice, and the ethics committee of the university hospital of Marburg. All patients gave informed and written consent.

Included: Patients with DCM n = 322

Examination for inflammatory/viral etiology on EMB available n = 272

Inflammatory/viral pos n = 119

Figure 1

Flow-chart of the study course.

Inflammatory/viral neg n = 153

4

5

8

18

Analysed n = 107

Analysed n = 130

Follow-up examination: 103 Telephone follow-up: 4

Follow-up examination: 120 Telephone follow-up: 10

Died n=9 Refused follow-up n = 26

European Journal of Clinical Investigation Vol 45

907

908

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation 74
6  9
7

77
5  10
4

Diastolic BP [mmHg]

30 (52)

2 (3)

III

IV

79
7  17
4

11 (19)

Heart rate [bpm]

Atrial fibrillation

ECG

13 (22)

19 (33)

II

Peripheral oedema

7 (12)

I

6 (9)

79
6  18
6

4 (6)

2 (3)

31 (48)

25 (38)

7 (11)

115
2  15
2

118
5  16
0

Systolic BP [mmHg]

NYHA class

26
2  4
5

27
6  4
9

10 (15%)

0 (0)

28 (43)

31 (48)

7 (11)

3 (5)

21 (32)

44
3  10
0

BMI [kg/m2]

Examination

9 (16%)

35 (60)

Hyperuricaemia

fDCM

31 (53)

Hyperlipidaemia

4 (7)

16 (21)

Hypertension

COPD

10 (17)

16 (28)

48
3  12
7

Diabetes mellitus

Risk factors

Female sex

Age [y]

Demographics

14 (19)

75
0  15
3

11 (15)

0 (0)

32 (44)

30 (42)

10 (14)

79
0  12
9

123
6  19
3

27
2  4
1

22 (31%)

9 (13)

31 (43)

44 (61)

21 (29)

14 (19)

16 (22)

56
3  11
5

9 (21)

74
5  13
8

7 (17)

2 (5)

17 (40)

20 (48)

3 (7)

75
9  10
3

119
9  16
8

29
2  3
8

12 (29%)

2 (5)

28 (67)

24 (57)

15 (36)

9 (21)

4 (10)

52
6  13
1

n = 42

n = 72

n = 58

No. of subjects

n = 65

Diagnosed > 1 year ago DCM DCMi

Diagnosed < 1 year ago DCM DCMi

Descriptive data Diagnosis: Aetiology:

0
185

0
912

0
044

0
482

0
943

0
930

0
003

0
234

0
002

0
902

0
023

0
052

0
083

0
924

DCM: OR = 1
08 (0
50 to 2
33) DCMi: OR = 4
60 (1
68 to 12
59)** OR = 1
41 (0
84 to 2
38) DCM: OR = 0
50 (0
25 to 1
00) DCMi: OR = 2
64 (1
18 to 5
93)* OR = 2
56 (0
79 to 8
26)

new: OR = 0
32 (0
12 to 0
84)* > 1y: OR = 1
35 (0
60 to 3
03) OR = 0
82 (0
49 to 1
38) new: OR = 0
50 (0
24 to 1
02) > 1y: OR = 2
65 (1
20 to 5
85)* OR = 0
21 (0
05 to 0
93)*

D = –4
8 (–9
1 to –0
5)* OR = 1
48 (0
75 to 2
94)

OR = 0
74 (0
37 to 1
48)

DCMi: OR = 3
05 (0
83 to 11
16)

> 1y: OR = 1
11 (0
39 to 3
12)

D = –0
3 (–4
6 to +4
0)

DCM: OR = 0
62 (0
26 to 1
52)

OR = 0
76 (0
47 to 1
25)

D = +1
4 (–1
4 to +4
3)

new: OR = 0
23/0
07 to 0
74)*

OR = 1
04 (0
64 to 1
71)

D = –3
0 (–5
9 to –0
1)*

D = +4
9 (+0
5 to +9
3)*

DCMi: D = +3
0 (+1
4 to +4
7)***

> 1y: D = +2
0 (+0
5 to +3
6)* D = –3
5 (–7
9 to +1
0)

DCM: D = –0
4 (–1
9 to +1
1)

new: D = –1
4 (–3
1 to +0
3)

OR = 2
31 (1
22 to 4
38)

OR = 1
93 (0
93 to 3
99)

OR = 0
48 (0
25 to 0
89)*

OR = 0
83 (0
44 to 1
54)

OR = 0
63 (0
30 to 1
33)

D = +8
1 (+5
1 to +11
2)***

D = –3
9 (–6
9 to –0
8)*

Differences between subgroups Interaction Difference, D/Odds Ratio, OR (95% CI) P-value DCMi vs. DCM > 1 year vs. < 1 year

Table 1 Baseline characteristics of patients with inflammatory (DCMi) vs. noninflammatory (DCM) aetiology, and with old (> 1 year) vs. new (< 1 year) diagnosis of heart failure

S. PANKUWEIT ET AL. www.ejci-online.com

67
1  7
8 42
7  8
4

139
0  15
3

65
9  6
8

44
4  6
9

135
6  14
2

LVDED [mm]

LADES [mm]

391  102 7
8  2
0

408  128

7
8  2
0

26 (45)

Uric acid [mmol/L]

Leucocytes [109/L]

CRP > 5 mg/L

15 (26)

17 (29)

Lipid lowering drug

Uric acid lowering drug

40 (62%)

7 (11)

10 (15)

47 (72)

48 (74)

58 (89)

63 (97)

4 (6%)

21 (29)

18 (25)

56 (78)

41 (57)

66 (92)

66 (92)

23 (32)

7
6  2
1

407  116

5
41  1
35

14
6  1
6

142
0  14
6

44
4  7
0

68
8  7
7

29
9  8
0

17 (40%)

15 (36)

7 (17)

37 (88)

27 (64)

37 (88)

40 (95)

19 (46)

8
5  2
6

444  105

5
34  1
47

14
8  2
2

140
4  13
7

48
1  6
8

69
5  6
8

29
8  8
4

Diagnosed > 1 year ago DCM DCMi

0
82

0
016

0
834

0
036

0
903

0
426

0
964

0
014

0
167

0
079

0
367

0
830

0
202

0
006

0
812

0
641

OR = 0
61 (0
35 to 1
06) DCM: OR = 0
64 (0
26 to 1
59) DCMi: OR = 2
83 (0
96 to 8
35) OR = 1
01 (0
53 to 1
90) DCM: OR = 0
99 (0
46 to 2
12) DCMi: OR = 4
60 (1
68 to 12
60)** OR = 0
45 (0
23 to 0
88) *

new: OR = 0
48 (0
20 to 1
17) > 1y: OR = 2
11 (0
71 to 6
27) OR = 0
56 (0
29 to 1
08) new: OR = 0
29 (0
11 to 0
77)** > 1y: OR = 1
35 (0
60 to 3
03) OR = 12
9 (6
1 to 27
4) ***

OR = 0
65 (0
25 to 1
68)

OR = 0
48 (0
18 to 1
25) OR = 1
32 (0
76 to 2
29)

OR = 0
61 (0
19 to 1
96)

DCMi: OR = 2
26 (0
99 to 5
12)

OR = 1
78 (0
52 to 6
03)

DCM: OR = 0
58 (0
28 to 1
18)

> 1y: OR = 1
84 (0
84 to 4
05)

D = +0
2 (–0
4 to +0
8)

D = +22 (–8 to +53)

D = +0
22 (–0
13 to +0
58)

D = +0
2 (–0
2 to +0
7)

new: OR = 0
47 (0
22 to 1
00)*

D = +0
4 (–0
2 to +1
0)

D = +8 (–23 to +38)

D = +0
11 (–0
25 to +0
47)

D = +0
3 (–0
1 to +0
8)

D = +4
2 (+0
4 to +8
0)*

DCMi: D = +5
4 (+2
4 to +8
5)***

D = +1
1 (–2
7 to + 4
9)

DCM: D = 0
0 (–2
4 to +2
4)

> 1y: D = +3
7 (+1
0 to +6
4)**

D = +2
8 (+0
8 to +4
6)**

D = –0
8 (–2
9 to +1
4)

new: D = –1
7 (–4
5 to +1
0)

D = +1
0 (–0
9 to +2
9)

D = –0
6 (–2
8 to +1
6)

Differences between subgroups Interaction Difference, D/Odds Ratio, OR (95% CI) P-value DCMi vs. DCM > 1 year vs. < 1 year

*P < 0
05; **P < 0
01; ***P < 0
001. COPD, chronic obstructive pulmonary disease; BMI, body mass index; BP, blood pressure; NYHA, New York Heart Association; QTc, corrected QT interval (Bazett’s formula); LVEF, left ventricular ejection fraction; LVDED, left ventricular end-diastolic diameter; LADES, left atrial end-systolic diameter; CRP, C-reactive protein; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; AldoRA, aldosterone receptor antagonist. Daily doses of ACEI/ARB and beta-blockers were computed as percent of maximum recommended daily doses of the respective agents.

6 (10%)

49 (85)

Immunoglobulin therapy

40 (69)

Diuretic

56 (97)

Beta-blocker

AldoRA

55 (95)

ACEI/ARB

Treatment

5
30  1
31

5
04  1
27

Total cholesterol [mmol/L]

18 (28)

14
6  1
5

14
2  1
6

Haemoglobin [g/dL]

Laboratory

Henry-Index [%]

30
0  8
5

31
1  8
6

Diagnosed < 1 year ago DCM DCMi

LVEF [%]

Echo-cardiography

Descriptive data Diagnosis: Aetiology:

Table 1 Continued

INFLUENCE OF DIFFERENT AETIOLOGIES

European Journal of Clinical Investigation Vol 45

909

S. PANKUWEIT ET AL.

Patient recruitment and follow-up Three-hundred and twenty-two stable, ambulatory patients of either sex with chronic heart failure demonstrated a left ventricular ejection fraction < 45% at baseline. All patients were diagnosed and treated as recommended by European and American guidelines valid during the study period [14–16]. Cardiac catheterization was performed in all patients to exclude coronary heart disease and an echocardiogram with assessment of a left ventricular ejection fraction < 45% and a Henry Index > 117% was performed in all patients. In addition to clinical data past medical history, cardiac and noncardiac diagnoses, current therapy, ECG and further relevant anamnestic data with respect to an inflammatory and/or viral aetiology were available in 272 patients from TP9a, in whom endomyocardial biopsy (EMB) was performed [5–7]. Investigation of the EMB included conventional histology, immunohistochemistry and molecular biology for the detection of cardiotropic viruses and was carried out as described previously [17,18]. In detail, immunohistochemistry was performed to demonstrate infiltrating cells by antibodies specific for activated T and B cells, macrophages, major histocompatibility class1 and class2 antigens, adhesion molecules and endothelial cells. Specific binding of the antibodies indicating an inflammatory reaction was demonstrated by peroxidase double staining procedure. Inflammation in endomyocardial biopsies was diagnosed by the presence of ≥ 14 lymphocytes/mm2. For demonstration of cardiotropic viruses to be present in the endomyocardial biopsy the QIAamp Tissue Kit (Qiagen, Hilden, Germany) was used to extract total DNA and RNA from the samples. Primer pairs specific for coxsackievirus B, parvovirus B19 (B19V), cytomegalovirus (CMV), adenovirus type 2, influenza virus A, human herpesvirus 6 (HHV6) and Epstein–Barr virus (EBV) were used to perform PCR and in case of B19V quantitative real-time PCR. PCR results were confirmed by southern blot hybridization. Depending on the results of investigation of the EMB a subgroup of patients was treated with immunoglobulins in addition to the heart failure treatment. Data with regard to the number of patients treated were given in Table 1. Patients on guideline-based medical therapy were followed for 12 months. Clinical evaluation, echocardiography and blood tests were performed during the follow-up visit in 223 patients from TP9a, whereas in 14 of these patients only telephone visits were carried out. For detailed analysis patients were divided into four subgroups with regard to the results of endomyocardial biopsy and time point of diagnosis: patients with DCM of nonviral and noninflammatory origin (DCM) and patients with DCM of inflammatory and/or viral origin (DCMi/v), dividing each subgroup into newly diagnosed (< 1 year) DCM or DCMi/v and DCM or DCMi/v diagnosed

910

www.ejci-online.com

later (more than 1 year ago and treated for longer than 1 year). (Fig. 1).

Echocardiography Echocardiography was performed by trained cardiologists, blinded to diagnosis and clinical treatment using a GE Vivid 7 ultrasound machine (3
5 MHz transducer) according to standard techniques as defined by the American Society of Echocardiography [19]. An ejection fraction (EF) < 45%, determined by Simpson’s monoplane method, was defined as systolic dysfunction. Furthermore, predicted LVEDD was calculated according to the formula of Henry: LVEDD = [45
3 9 body surface area 0
3] [0
03 9 age] 7
2. All DCM patients included in the investigation have had a predicted LVEDD > 117% [20]. None of the patients had significant primary valvular disease or known relevant coronary artery stenosis. Additional imaging methods as MRI or CT were not performed in the patients.

Statistical analysis Patients were analysed in four groups distinguished by time of first diagnosis of heart failure (new (< 1 year) vs. later (> 1 year ago)) and aetiology (noninflammatory DCM vs. inflammatory DCMi/v). Groupwise baseline data are presented as frequencies (percent) or mean  standard deviation. Between-group comparisons were carried out by binary or ordinal logistic regression or two-way analysis of variance as appropriate. Results were presented as odds ratios or mean differences with 95% confidence intervals. First, an interaction test was performed to examine whether the effects of duration of heart failure and aetiology were independent from each other. In case of a nonsignificant interaction, overall estimates for the differences between < 1 year and > 1 year diagnosis, and between DCM and DCMi/v were provided. If an interaction was detected, the effect of time of diagnosis was estimated separately for DCM and DCMi/v and, vice versa, the difference between DCM and DCMi/v was estimated separately in new (< 1 year) and old (> 1 year) diagnosis. Outcomes at the follow-up were evaluated by the same methods, including the respective baseline values as covariates. If the outcome was found to be independent of time since diagnosis and aetiology, a pooled estimate for the change from baseline to follow-up and its 95% confidence interval was computed. In case of dependency on only one of both factors, separate estimates for the change from baseline were provided in the corresponding two subgroups. In case of dependency on both factors (including presence of interaction), changes from baseline were estimated separately in all four subgroups. New York Heart Association class is a categorical variable. For an easier visibility we used mean values for graphical

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation

total: –5
9 (–8
3 to –3
4)***

Change (95% CI)

new: –5
5 (–7
0 to –4
0)***

Change (95% CI)

new: –3
3 (–4
7 to –1
8)***

Change (95% CI)

–0
82 (–1
05 to –0
58)***

–0
48 (–0
71 to –0
26)***

Change (95% CI)

35
0  9
4

73
1  12
9

76
0  11
0

120
2  17
7

29
4  4
0

2
26  0
66 –0
17 (–0
38 to +0
05)

–0
24 (–0
38 to –0
10)**

47
6  7
1

68
1  8
9

2
07  0
72

> 1y: +0
1 (–1
0 to +1
2)

45
3  7
8

> 1y: –1
8 (–3
0 to –0
5)**

67
1  9
6

> 1y: +5
8 (+3
6 to +8
0)***

35
8  11
6

70
1  16
0

> 1y: –1
5 (–4
1 to +1
2)

77
6  11
0

> 1y: –1
4 (–5
2 to +2
4)

122
1  18
6

27
4  4
3

Diagnosed > 1 year ago DCM DCMi

0
002

0
286

0
165

0
182

0
186

0
587

0
571

0
713

> 1y: OR = 1
61 (0
71 to 3
66)

new: OR = 0
35 (0
18 to 0
71)**

D = –0
6 (–2
1 to +1
0)

D = –0
7 (–2
7 to +1
2)

D = +1
6 (–1
6 to +4
8)

D = +0
5 (–3
2 to + 4
2)

D = +0
1 (–2
8 to +3
0)

D = –1
6 (–6
4 to +3
2)

D = +0
3 (–0
2 to +0
8)

DCMi: OR = 5
93 (2
62 to 13
41)***

DCM: OR = 1
48 (0
73 to 2
98)

D = +4
2 (+2
6 to +5
8)

D = +4
1 (+2
1 to +6
1)***

D = –7
8 (–11
0 to –4
7)***

D = +1
8 (–1
9 to +5
5)

D = –3
4 (–6
3 to –0
5)*

D = –5
6 (–10
4 to –0
8)*

D = –0
4 (–0
9 to +0
1)

Differences between subgroups Interaction Difference, D/Odds Ratio, OR (95% CI) P-value DCMi vs. DCM > 1 year vs. new

*P < 0
05; **P < 0
01; ***P < 0
001. BMI, body mass index; BP, blood pressure; QTc, corrected QT interval; LVEF, left ventricular ejection fraction; LVDED, left ventricular end-diastolic diameter; LADES, left atrial end-systolic diameter; NYHA, New York Heart Association.

1
62  0
72

1
98  0
71

39
5  6
1

Follow-up

NYHA class

41
9  6
0

Follow-up

LADES [mm]

61
9  7
7

Follow-up

61
1  10
5

new: +13
6 (+11
2 to +16
1)***

Change (95% CI)

LVDED [mm]

42
0  13
7

45
5  14
1

69
8  12
3

80
0  10
9

Follow-up

LVEF [%]

71
7  14
3

Follow-up

Heart rate [bpm]

79
7  11
7

new: +4
0 (+1
5 to +6
4)**

Follow-up

Change (95% CI)

Diastolic BP [mmHg]

new: +7
7 (+3
9 to +11
5)***

Change (95% CI)

122
4  19
9

126
5  19
5

Follow-up

Systolic BP [mmHg]

28
2  4
8

Total: +0
5 (+0
3 to +0
8)***

27
2  4
3

DCMi

Follow-up

Diagnosed < 1 year ago DCM

Change (95% CI)

BMI [kg/m2]

Descriptive data Diagnosis: Aetiology:

Table 2 Follow-up characteristics and changes from baseline to follow-up of patients with inflammatory vs. noninflammatory aetiology, and with old (> 1 year) vs. new (< 1 year) diagnosis of heart failure

INFLUENCE OF DIFFERENT AETIOLOGIES

European Journal of Clinical Investigation Vol 45

911

S. PANKUWEIT ET AL.

illustration. However, to optimize accuracy an ordinal logistic regression analysis was performed (Table 3). All analyses were performed by SPSS version 18 (Armonk, NY, USA).

Results Study subjects Of 322 patients presenting with dilated cardiomyopathy, 272 patients underwent endomyocardial biopsy for the assessment of an inflammatory aetiology. With regard to the investigation of endomyocardial biopsy in 17 (26%) of patients with DCMi diagnosed < 1 year ago an inflammation, in 32 (49%) patients virus genome and in 16 (25%) patients virus genome and an inflammation were found. In 23 (55%) of patients with DCMi diagnosed longer than year before an inflammation, in 10 (24%) patients virus genome and in nine (21%) patients virus genome and an inflammation were found. In summary, we were able to identify a viral and/or inflammatory aetiology in 107 patients with B19V as the most often detected virus. In only one patient of each subgroup HHV6 was detected instead of B19V. A double infection with P19V and HHV6 was diagnosed in only one patient within the group DCMi diagnosed < 1 year ago. Nine patients died within 1 year, five patients because of advanced heart failure, two patients because of malignant disease and two patients because of sudden cardiac death. Twenty-six patients refused follow-up. Data of the remaining 237 patients at baseline in accordance to the different subgroups are presented in Table 1. A one-year-follow-up complete clinical examination was performed in 223 patients. Additional fourteen subjects only provided clinical information (general health, medication, NYHA class, body weight and results of echocardiogram) by telephone assessment. In addition, characteristics of patients diagnosed with DCM or DCMi/ v more than 1 year ago are demonstrated in Table 1.

Baseline data The number of subjects in the four subgroups and baseline characteristics are shown in Table 1. Patients who were later diagnosed with heart failure were older than those who were newly diagnosed. Independent from the time of diagnosis, subjects with DCM were older than those with DCMi/v. Among the newly diagnosed cases more patients were female. Compared to patients with newly diagnosed heart failure, those patients who had a history of heart failure of 1 year and more were less frequently smokers, reported more often on a history of syncope, suffered from a higher blood pressure (significant for systolic), showed a lower heart rate, a longer QRS interval and a larger left ventricle and were more often diagnosed with fDCM. They were more often treated with betablockers and overall carried more often an implantable device.

912

www.ejci-online.com

Compared to patients with DCM, patients with DCMi/v suffered less frequently from chronic obstructive pulmonary disease and lower blood pressure (significant for diastolic). Relationships between pathogenesis (inflammation/viral vs. no inflammation/nonviral), duration of heart failure and clinical course of patients were dependent from each other, as shown by multiple statistical interactions. Newly diagnosed DCMi/v presented with the more favourable, and later diagnosed DCMi/v with less favourable characteristics such as prevalence of hypertension and hyperuricaemia, body mass index, left atrial size, abnormal CRP, and use of uric acid lowering drugs. Peripheral oedema was less frequent in newly diagnosed DCMi/v unlike in DCM, where the newly diagnosed cases had the highest prevalence. An opposite pattern was observed with respect to mitral regurgitation which was most prevalent in newly diagnosed DCMi/v and least frequent in later diagnosed DCMi/v. However, after adjusting for age and sex, in a part of the variables described above formal significances were lost while trends were preserved. These losses of significance were caused by merely small shifts of the P-values across the margin of 0
05. Further analyses revealed that adjustment for age but not for sex caused these changes.

Changes from baseline to one-year follow-up One-year-follow-up data are shown in Table 2. Body mass index increased from baseline to follow-up in all patients without significant differences between groups. In the same time, blood pressure increased in newly diagnosed cases but remained without significant changes in the other subgroups of patients. Heart rate was lowered unanimously in all four subgroups, but QRS and QTc intervals were smaller only in newly diagnosed patients while they remained without significant changes in the other subgroups. Similary, there was an improvement in LVEF and left ventricular size which was significantly stronger in the newly diagnosed subgroup. Moreover, a reduction in left atrial size was observed only in this subgroup. Furthermore, we could demonstrate a comparable modest improvement of NYHA class in both DCM and DCMi/v patients with later diagnosed DCM, but a significantly stronger improvement in newly diagnosed DCM with an extra benefit in newly diagnosed DCMi/v patients (P < 0
001 for DCMi/v vs. DCM in patients with newly diagnosed DCM). (Fig. 2, Tables 2 and 3) These effects were not altered when adding immunoglobulin therapy to the analyses. In addition, as demonstrated in Table 2, changes in NYHA class were most favourable in newly diagnosed DCMi/v (P < 0
001 vs. each of the three other groups) and did not differ significantly between the other subgroups (P = 0
311 and 0
056 for newly diagnosed DCM vs.

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation

INFLUENCE OF DIFFERENT AETIOLOGIES

Inflammatory aetiology no

yes

2·6

NYHA class Baseline

Mean +/– SEM

2·4

Follow-up

2·2

2·0

1·8

1·6

Figure 2 Changes of New York Heart Association class from baseline to followup, depending on aetiology (inflammatory/ viral) and time since diagnosis of heart failure.

1·4 Newly diagnosed

>1 year ago

Newly diagnosed

>1 year ago

First diagnosis of heart failure

Table 3 Changes in New York Heart Association class from baseline to follow-up of patients with inflammatory vs. non-inflammatory aetiology, and with old (> 1 year) vs. new (< 1 year) diagnosis of heart failure Diagnosis: Aetiology:

Diagnosed < 1 year ago DCM

Diagnosed < 1 year ago DCMi

Diagnosed > 1 year ago: DCM

Diagnosed > 1 year ago DCMi

Improved

25 (43%)

40 (62%)

19 (26%)

12 (29%)

Unchanged

28 (48%)

22 (34%)

49 (68%)

24 (57%)

5 (9%)

3 (4%)

4 (6%)

6 (14%)

20% (8–33)**

15% (( 5) 34)

Worsened Net gain (95% CI)

34% (18–51)***

58% (42–71)***

**P < 0
01; ***P < 0
001.

both later diagnosed subgroups, P = 0
282 between later diagnosed subgroups). This justifies to model the effect of newly diagnosed DCMi/v vs. all other patients. Changes in ACEI/ ARB doses, expressed as per cent of the recommended daily target dose, were significantly different in the four groups (P < 0
001): +5 (95%CI 2 to +12) in later diagnosed DCM, 6 ( 17 to +5) in later diagnosed DCMi/v, +17 (+9 to +25) in newly diagnosed DCM, and +17 (+10 to +24) in newly diagnosed DCMi/v, where both ‘newly diagnosed’ groups differed significantly from both ‘later diagnosed’ groups, and both later diagnosed groups and both newly diagnosed groups were comparable respectively. However, these differences did not explain the between-group differences of changes in NYHA

class. In addition, Table 4 demonstrates that also younger age and female sex were associated with an early improvement in NYHA class (P = 0
004 for age and 0
007 for female sex respectively). No significant between-group differences of changes in beta-blocker and aldosterone receptor blocker prescriptions could be found. In Fig. 3 changes in LVEF from baseline to follow-up are demonstrated. LVEF improved in all four subgroups. The gain was strongest in patients with newly diagnosed heart failure and especially in patients with newly diagnosed DCMi/v. But next to NYHA class, this trend for interaction was similar, however not statistically significant with regard to an improvement of LVEF in patients with DCMi/v.

European Journal of Clinical Investigation Vol 45

913

S. PANKUWEIT ET AL.

www.ejci-online.com

Table 4 Predictors of higher NYHA class at one-year follow-up in simple and multiple ordinal logistic regression Simple regression*

Multiple regression Stepwise backward elimination‡

Full model† Variable

Odds Ratio (95%CI)

P-value

Odds Ratio (95%CI)

P-value

Odds Ratio (95%CI)

P-value

Inflammatory/viral aetiology and new diagnosis of heart failure

0
21 (0
12 to 0
39)

< 0
001

0
26 (0
13 to 0
50)

< 0
001

0
25 (0
13 to 0
48)

< 0
001

Age (per +10 years)

1
58 (1
29 to 1
95)

< 0
001

1
41 (1
12 to 1
79)

0
004

1
34 (1
07 to 1
68)

0
010

Female sex

1
85 (1
03 to 3
31)

0
038

2
62 (1
30 to 5
30)

0
007

2
77 (1
41 to 5
46)

0
003

Body mass index (per +1 kg/m2)

1
05 (0
99 to 1
11)

0
115

1
04 (0
97 to 1
11)

0
271

–

–

Systolic blood pressure (per +10 mmHg)

1
07 (0
92 to 1
23)

0
391

0
92 (0
78 to 1
08)

0
313

–

–

Heart rate (per +10 bpm)

0
84 (0
72 to 0
98)

0
029

0
83 (0
70 to 0
99)

0
036

0
86 (0
73 to 1
02)

Left ventricular ejection fraction (per +10%)

0
90 (0
67 to 1
22)

0
517

0
86 (0
60 to 1
22)

0
383

–

End-systolic left atrial diameter (per +5 mm)

1
12 (0
95 to 1
33)

0
177

1
17 (0
92 to 1
78)

0
194

1
25 (1
02 to 1
52)

Mitral regurgitation

1
16 (0
69 to 1
97)

0
574

1
02 (0
55 to 1
89)

0
960

–

–

Uric acid (per +100 mmol/L)

1
02 (0
86 to 1
21)

0
823

0
93 (0
77 to 1
14)

0
498

–

–

CRP > 5 mg/L

1
61 (0
96 to 2
71)

0
070

1
53 (0
86 to 2
70)

0
146

–

–

ACE inhibitor/angiotensin receptor blocker

1
13 (0
39 to 3
30)

0
818

1
72 (0
54 to 5
49)

0
360

–

–

Beta-blocker

0
39 (0
17 to 0
95)

0
038

0
32 (0
12 to 0
88)

0
027

0
35 (0
13 to 0
92)

Peripheral oedema

2
07 (1
01 to 4
25)

0
049

1
13 (0
51 to 2
49)

0
770

–

0
074 – 0
029

0
034 –

NYHA, New York Heart Association; CRP, C-reactive protein; ACE, angiotensin-converting enzyme. *Effect of each variable adjusted for baseline NYHA class. † Including all variables simultaneously, adjusting additionally for baseline NYHA class. ‡ Starting with full model, stepwise excluding the most nonsignificant variable until P ≤ 0
10 for all remaining variables.

Furthermore, with respect to familial DCM we calculated that 9 (16%) patients in newly diagnosed DCM and 22 (31%) in later diagnosed DCM suffered from fDCM (DCMi/v new 10 (15%), later diagnosed 12 (29%), Odds Ratio for later diagnosed vs. new 2
31 (1
22–4
38), P = 0
011). Family history indicative of DCM did not reach significance when included into the follow-up analyses in Table 2, nor did its inclusion alter any of the effects reported in Table 2.

Discussion As a major result of our analysis we could demonstrate a significant interaction between duration of disease, NYHA class and left ventricular ejection fraction in patients with DCM. A significant improvement in NYHA class especially in patients with newly diagnosed DCMi/v when compared to patients who suffer from DCM could be described. The clinical improvement is accompanied by an increase in LVEF and a reduction in LVEDD. Furthermore it has been shown that the improvement of NYHA class depends on the time-point of the initial diagnosis. In case of a later diagnosis (> 1 year) of

914

disease the improvement in NYHA class is smaller. These effects are independent from an immunoglobulin treatment especially in a subgroup of patients with DCMi/v. The results are in accordance to the results published by McNamara et al. in 2011, who demonstrated a substantial improvement in LVEF for most patients with recent onset of dilated cardiomyopathy over the first 6 months. In their investigation NYHA functional class IV and black race were associated with a significantly higher risk of death/transplantation over a 4 years follow-up. Female sex was associated with a significantly lower risk of death/transplantation [21]. But of interest, in the newly diagnosed group more patients were female and the mean age of patients was below the mean age of patients in the group, in which myocarditis had been diagnosed more than 1 year before. But these data are at least in part in accordance with results of other studies. This might be a recruitment bias, however, another possible explanation might be that hormonal factors may play a role in the aetiopathogenesis of myocardial inflammatory syndromes and that the process of myocardial remodelling depends on the course of inflammation [22–24].

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation

INFLUENCE OF DIFFERENT AETIOLOGIES

Inflammatory aetiology no

yes

50

LVEF Baseline

Follow-up

Mean +/– SEM

45

40

35

30

Figure 3 Changes of left ventricular ejection fraction from baseline to follow-up, depending on aetiology (inflammatory/ viral) and time since diagnosis of heart failure.

25 Newly diagnosed

Patients in our study cohort all suffered from inflammatory/viral and noninflammatory/nonviral DCM, in which we were able to differentiate between inflammatory-viral and inflammatory-nonviral aetiology. About half of the patients suffered from an inflammatory aetiology with or without the detection of viral genome (in most cases Parvovirus B19), or detection of viral genome without an inflammatory response. However, limited data on the causes of myocarditis show that viral infections are among the most common causes in DCM in developed countries [25]. Furthermore it has been demonstrated that virus-induced myocyte damage may lead to the release of intracellular proteins that trigger immunological responses in the presence of a predisposing genetic background. This mechanism may result in an ongoing post-viral myocardial damage even in the absence of residual viral genome at a later stage of disease and inflammation [26–28]. With regard to the genetic background and predisposing factors, the hypothesis of genetically driven, inflammatory mechanisms in DCM was recently supported by the identification of SNPs located within the MHC region on chromosome 6 using genome-wide association studies (GWAS) [29,30]. Genetic variants might modify individual susceptibility to DCM mediated by alterations of autoimmunity as well as immune competency against viruses, they might promote viral persistence in the myocardium and at the end might

>1 year ago

Newly diagnosed

>1 year ago

First diagnosis of heart failure response to anti-inflammatory therapies. For that reason, GWAS as well as next generation sequencing with high analytical quality and feasibility as shown by Haas et al. [31] may identify further genetic variants inducing inflammatory mechanisms as important pathophysiological pathways in heart failure progression. In a subanalysis we could demonstrate that with respect to the results demonstrated above no differences did exist even when calculated with an inflammatory-viral or inflammatorynonviral aetiology of DCM. Probably the number of patients in our study is too low to get further valid information about these two subgroups. An enlarged and possibly multicenter study is required to give an answer to this question, particularly on the condition of a clarified pathophysiology. Another important point of discussion is related to the therapy of these patients. Supportive care as a first line treatment has been applied to all patients. This includes heart failure therapy as recommended by the European Society of Cardiologists and the American Heart Association [14–16]. With regard to myocardial inflammation, an immunologic activation due to initial direct myocardial invasion by cardiotropic virus is discussed in human myocarditis. For that reason, all patients in our study with detection of viral genome with or without signs of inflammation in the endomyocardial biopsy were treated with a polyvalent intravenous immunoglobulin preparation. In

European Journal of Clinical Investigation Vol 45

915

S. PANKUWEIT ET AL.

general only recommendations with regard to medical therapy in addition to conventional heart failure treatment exist that depends on histopathological findings outside of randomized studies [32]. For this reason no guidelines with regard to this topic exist. On the one hand, there a few studies with regard to intravenous immuno globulins (IvIG) treatment in patients with myocarditis or recent onset of dilated cardiomyopathy. In 1994 Drucker et al. [33] reported a series of children treated with IvIg in whom significant improvement of left ventricular function was seen when compared with historical controls. Two uncontrolled series of adults with recent-onset dilated cardiomyopathy and peripartum cardiomyopathy treated with IvIg had substantial recovery of left ventricular function during follow-up [34,35]. On the other hand, there are data from a controlled trial of immunoglobulin-treatment in patients with recent onset cardiomyopathy published by McNamara in 2001. In this cohort, LVEF improved significantly during follow-up, and the short-term prognosis remains favourable, but IVIG does not contribute to the improvement of LVEF. However, as a limitation of this study, inflammation was only confirmed by histological investigation of the endomyocardial biopsy. Methods with higher sensitivity for the detection of viral or inflammatory heart disease as immunohistochemistry or molecular biology for the detection of cardiotropic viruses were not used [21]. Our results clearly demonstrate that in DCM and especially in DCMi/v early diagnosis is associated with a strong improvement in the clinical course, whereas a long duration of dilated cardiomyopathy results in a loss of changes in the clinical course and outcome [36].

Conclusion The improvement in NYHA class, LVEF and a reduction in LVEDD in patients with DCM depends on the time course of the disease: best results with respect to clinical parameters can be obtained in patients with a short duration of disease and in patients who suffer from DCMi/v. These important epidemiological data emphasize the high relevance of an early diagnosis to start heart failure treatment as soon as possible with probably positive effects on cardiac remodelling especially in patients with DCMi/v. Nevertheless, assessment of pathogenesis by histopathology, aetiology and molecular findings obviously bears additional prognostic impact. Acknowledgements This study was supported by the German Competence Network Heart Failure, TP9, FKZ 01GI0205. Competing interests The authors declare that they have no competing interests.

916

www.ejci-online.com

Author contributions SP participated in the design of the study, carried out the immunoassays and helped to draft the manuscript. CL participated in the design of the study and drafted the manuscript. AR participated in the design of the study. VR participated in the design of the study and carried out the immunoassays. GG performed the statistical analysis. BM participated in the design of the study and helped to draft the manuscript. Address Department of Internal Medicine – Cardiology Angiology Intensive Care and Prevention, University of Marburg, Marburg, Germany (S. Pankuweit, A. Richter, V. Ruppert); Department of Cardiology, Klinikum Oldenburg, University of Oldenburg, European Medical School, Oldenburg, Germany (C. L€ uers); Institute of Clinical Epidemiology and Biometry (ICEB), Julius Maximilian University of W€ urzburg, W€ urzburg, Germany (G. Gelbrich); Herz- und Gef€ aßzentrum Marburg (HGZ), Marburg, Germany (B. Maisch). Correspondence to: Claus Luers, MD, Department of Internal Medicine – Cardiology, University of Oldenburg, European Medical School, Rahel-Straus-Str. 10, 26133 Oldenburg, Germany. Tel.: +49-441-40377155; fax: +49-441-4032784; e-mail: [email protected] Received 7 December 2014; accepted 17 June 2015 References 1 Burkett EL, Hershberger RE. Clinical and genetic issues in familial dilated cardiomyopathy. J Am Coll Cardiol 2005;45:969–81. 2 Dec GW, Fuster V. Idiopathic dilated cardiomyopathy. N Engl J Med 1994;331:1564–75. 3 Mestroni L, Maisch B, McKenna WJ, Schwartz K, Charron P, Rocco C et al. Guidelines for the study of familial dilated cardiomyopathies. Collaborative Research Group of the European Human and Capital Mobility Project on Familial Dilated Cardiomyopathy. Eur Heart J 1999;20:93–102. 4 Elliott P, Andersson B, Arbustini E, Bilinska Z, Cecchi F, Charron P et al. Classification of the cardiomyopathies: a position statement from the European Society Of Cardiology Working Group on Myocardial and Pericardial Diseases. Eur Heart J 2008;29:270–6. 5 Maisch B, Noutsias M, Ruppert V, Richter A, Pankuweit S. Cardiomyopathies: classification, diagnosis, and treatment. Heart Fail Clin 2012;8:53–78. 6 Maisch B, Richter A, Sandmoller A, Portig I, Pankuweit S. Inflammatory dilated cardiomyopathy (DCMI). Herz 2005;30:535–44. 7 Pankuweit S, Richter A, Ruppert V, Funck R, Maisch B. [Classification, genetic predisposition and risk factors for the development of cardiomyopathies]. Der Internist 2008;49:441–2, 444–7. 8 Kuhl U, Pauschinger M, Schwimmbeck PL, Seeberg B, Lober C, Noutsias M et al. Interferon-beta treatment eliminates cardiotropic viruses and improves left ventricular function in patients with

ª 2015 Stichting European Society for Clinical Investigation Journal Foundation

INFLUENCE OF DIFFERENT AETIOLOGIES

9

10 11 12

13

14

15

16

17

18

19

20

myocardial persistence of viral genomes and left ventricular dysfunction. Circulation 2003;107:2793–8. Westermann D, Lindner D, Kasner M, Zietsch C, Savvatis K, Escher F et al. Cardiac inflammation contributes to changes in the extracellular matrix in patients with heart failure and normal ejection fraction. Circ Heart Fail 2011;4:44–52. Report of the WHO/ISFC task force on the definition and classification of cardiomyopathies. Br Heart J 1980;44:672–3. Luk A, Ahn E, Soor GS, Butany J. Dilated cardiomyopathy: a review. J Clin Pathol 2009;62:219–25. Maisch B, Portig I, Ristic A, Hufnagel G, Pankuweit S. Definition of inflammatory cardiomyopathy (myocarditis): on the way to consensus. A status report. Herz 2000;25:200–9. Richardson P, McKenna W, Bristow M, Maisch B, Mautner B, O’Connell J et al. Report of the 1995 World Health Organization/ International Society and Federation of Cardiology Task Force on the Definition and Classification of cardiomyopathies. Circulation 1996;93:841–2. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG et al. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: endorsed by the Heart Rhythm Society. Circulation 2005;112:e154–235. Jessup M, Abraham WT, Casey DE, Feldman AM, Francis GS, Ganiats TG et al. 2009 focused update: ACCF/AHA Guidelines for the Diagnosis and Management of Heart Failure in Adults: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation 2009;119:1977–2016. Remme WJ, Swedberg K. Task Force for the D and Treatment of Chronic Heart Failure ESoC. Guidelines for the diagnosis and treatment of chronic heart failure. Eur Heart J 2001;22: 1527–60. Caforio AL, Pankuweit S, Arbustini E, Basso C, Gimeno-Blanes J, Felix SB et al. Current state of knowledge on aetiology, diagnosis, management, and therapy of myocarditis: a position statement of the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases. Eur Heart J 2013;34:2636–48. Pankuweit S, Ristic AD, Seferovic PM, Maisch B. Bacterial pericarditis: diagnosis and management. Am J Cardiovasc Drugs 2005;5:103–12. Cheitlin MD, Armstrong WF, Aurigemma GP, Beller GA, Bierman FZ, Davis JL et al. ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography–summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). J Am Coll Cardiol 2003;42:954–70. Henry WL, Gardin JM, Ware JH. Echocardiographic measurements in normal subjects from infancy to old age. Circulation 1980;62:1054–61.

21 McNamara DM, Holubkov R, Starling RC, Dec GW, Loh E, TorreAmione G et al. Controlled trial of intravenous immune globulin in recent-onset dilated cardiomyopathy. Circulation 2001;103:2254–9. 22 Buiatti A, Merlo M, Pinamonti B, De Biasio M, Bussani R, Sinagra G. Clinical presentation and long-term follow-up of perimyocarditis. J Cardiovasc Med 2013;14:235–41. 23 Imazio M, Brucato A, Barbieri A, Ferroni F, Maestroni S, Ligabue G et al. Good prognosis for pericarditis with and without myocardial involvement: results from a multicenter, prospective cohort study. Circulation 2013;128:42–9. 24 Imazio M, Trinchero R. Myopericarditis: Etiology, management, and prognosis. Int J Cardiol 2008;127:17–26. 25 Imazio M, Cecchi E, Demichelis B, Chinaglia A, Ierna S, Demarie D et al. Myopericarditis versus viral or idiopathic acute pericarditis. Heart 2008;94:498–501. 26 Wessely R, Henke A, Zell R, Kandolf R, Knowlton KU. Low-level expression of a mutant coxsackieviral cDNA induces a myocytopathic effect in culture: an approach to the study of enteroviral persistence in cardiac myocytes. Circulation 1998;98:450–7. 27 Kandolf R, Klingel K, Zell R, Canu A, Fortmuller U, Hohenadl C et al. Molecular mechanisms in the pathogenesis of enteroviral heart disease: acute and persistent infections. Clin Immunol Immunopathol 1993;68:153–8. 28 Klingel K, Hohenadl C, Canu A, Albrecht M, Seemann M, Mall G et al. Ongoing enterovirus-induced myocarditis is associated with persistent heart muscle infection: quantitative analysis of virus replication, tissue damage, and inflammation. Proc Natl Acad Sci USA 1992;89:314–8. 29 Meder B, Ruhle F, Weis T, Homuth G, Keller A, Franke J et al. A genome-wide association study identifies 6p21 as novel risk locus for dilated cardiomyopathy. Eur Heart J 2014;35:1069–77. 30 Villard E, Perret C, Gary F, Proust C, Dilanian G, Hengstenberg C et al. A genome-wide association study identifies two loci associated with heart failure due to dilated cardiomyopathy. Eur Heart J 2011;32:1065–76. 31 Haas J, Frese KS, Peil B, Kloos W, Keller A, Nietsch R et al. Atlas of the clinical genetics of human dilated cardiomyopathy. Eur Heart J 2015;36:1123–35. 32 Maisch B, Hufnagel G, Schonian U, Hengstenberg C. The European Study of Epidemiology and Treatment of Cardiac Inflammatory Disease (ESETCID). Eur Heart J 1995;16(Suppl O):173–5. 33 Drucker NA, Colan SD, Lewis AB, Beiser AS, Wessel DL, Takahashi M et al. Gamma-globulin treatment of acute myocarditis in the pediatric population. Circulation 1994;89:252–7. 34 Bozkurt B, Villaneuva FS, Holubkov R, Tokarczyk T, Alvarez RJ Jr, MacGowan GA et al. Intravenous immune globulin in the therapy of peripartum cardiomyopathy. J Am Coll Cardiol 1999;34:177–80. 35 McNamara DM, Rosenblum WD, Janosko KM, Trost MK, Villaneuva FS, Demetris AJ et al. Intravenous immune globulin in the therapy of myocarditis and acute cardiomyopathy. Circulation 1997;95:2476–8. 36 Cooper LT, Baughman KL, Feldman AM, Frustaci A, Jessup M, Kuhl U et al. The role of endomyocardial biopsy in the management of cardiovascular disease: a scientific statement from the American Heart Association, the American College of Cardiology, and the European Society of Cardiology. Endorsed by the Heart Failure Society of America and the Heart Failure Association of the European Society of Cardiology. J Am Coll Cardiol 2007;50:1914–31.

European Journal of Clinical Investigation Vol 45

917