www.elsevier.com/locate/heafai

The European Journal of Heart Failure 6 (2004) 723–729

Clinico-pathological evaluation of restrictive cardiomyopathy

(endomyocardial fibrosis and idiopathic restrictive cardiomyopathy)

in India

Sandeep Setha, Deepak Thataia, Sanjeev Sharmab, Prem Choprac, K.K. Talwara,*

aDepartments of Cardiology, Cardiothoracic Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110 029, IndiabDepartments of Cardiac Radiology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110 029, India

cDepartment of Pathology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110 029, India

D

ow

nl

Received 4 February 2003; received in revised form 19 June 2003; accepted 25 November 2003

Available online 18 August 2004

oaded

by guest on June 5, 2013

http://eurjhf.oxfordjournals.org/ from

Abstract

Background: Restrictive heart disease is characterized by impairment of ventricular filling during diastole with preserved systolic

function. The clinical and histopathological profile on endomyocardial biopsy of a cohort of patients with restrictive cardiomyopathy (RCM)

is presented.Methodology: The medical records of patients presenting with heart failure with systemic congestion, subsequently diagnosed as

restrictive heart disease after evaluation including cardiac catheterisation, were studied retrospectively to determine the clinical spectrum of

restrictive cardiomyopathy. The diagnosis of RCM was made, based on systemic congestion with dilated atria and near normal ventricular

size and function. Only patients who had an endomyocardial biopsy were included in the study. Patients with chronic constrictive pericarditis

and secondary restrictive heart disease mainly amyloidosis were excluded from the study. Results: All 52 patients had heart failure with

normal or near normal left ventricular size and function. Based on right and left ventricle angiography, patients were classified into two

groups. Group I with findings suggestive of EMF (n = 30) and Group II no evidence of EMF on angiography i.e. ‘idiopathic RCM’ (IRCM)

(n = 22). Baseline characteristics were similar in the two groups. Echocardiography revealed typical features of endomyocardial fibrosis in

Group I patients, with apical obliteration of right and left ventricular apices. Group II patients had no apex obliteration (except in four

patients, who were misclassified and in whom angiography did not show apex obliteration). The Group II patients had features of IRCM in

the form of normal left and right ventricular size and function with restrictive features of doppler filling along with dilated left and right atria.

Angiocardiography in EMF patients showed isolated RV involvement in only two patients. In the remaining 28 patients, the obliterative

changes were biventricular with RV involvement more severe than LV involvement. Angiographic findings in Group II (IRCM) patients were

unremarkable with preservation of normal trabecular pattern and absence of obliterative changes. Mild atrioventricular regurgitation was

present in 10/22 patients. Histopathological examination revealed that endocardial thickening was more common (77% vs. 23%) in EMF

patients. The presence of myocyte hypertrophy (70–80%), myocytolysis (40–50%) and interstitial fibrosis (46–56%) were similar in both

groups. Conclusions: The majority of our patients had biventricular EMF. A significant number of patients had clinical hemodynamic

features of restrictive heart disease but no evidence of EMF on angiography. These IRCM patients had similar clinical profiles to EMF but on

endomyocardial biopsy the endocardial thickening was minimal and seen in few patients (5/22).

D 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.

Keywords: Endomyocardial fibrosis; Pathology; Idiopathic restrictive cardiomyopathy

1. Introduction

Restrictive heart disease, which is characterized by

impairment of ventricular filling during diastole with

1388-9842/$ - see front matter D 2003 European Society of Cardiology. Publishe

doi:10.1016/j.ejheart.2003.11.009

* Corresponding author. Tel.: +91-1126594681; fax: +91-1126862663.

E-mail address: kktalwar@hotmail.com (K.K. Talwar).

preserved systolic function, may be caused by disorder

of any of the three layers of the heart [1]. Diastolic filling

in chronic constrictive pericarditis is restricted by the rigid

inelastic pericardium with indistinguishable visceral and

parietal layers [2]. Restrictive cardiomyopathies (RCM)

that result in decreased myocardial compliance include

those associated with infiltrative disorders namely amy-

loidosis, sarcoidosis and hemochromatosis [3–5]. In endo-

d by Elsevier B.V. All rights reserved.

Table 1

Clinical characteristics of both groups (meanF S.D.) along with ECG and chest X-ray findings (number (%))

Age Males Symptoms Functional class ECG CX-r

(Yrs) Duration II III IV AF RAE LAE CTR PVH

(months)

EMF 24F 12 19 20F 21 6 20 4 5 13 8 0.67F 0.10 15

(n= 30) (43%) (28%)

IRCM 26.5F 15 17 19F 20 5 13 4 4 10 7 0.61F 0.01 13

(n= 22) (45%) (31%)

AF atrial fibrillation, RAE right atrial enlargement, LAE left atrial enlargement, CTR cardiothoracic ratio, PVH pulmonary venous hypertension.

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729724

http://eurjhf.oxfoD

ownloaded from

myocardial fibrosis (EMF), the hemodynamic alterations

are mainly due to thickened and fibrotic endocardium

[6,7]. Idiopathic restrictive cardiomyopathy (IRCM) is

characterized by the clinical syndrome of restrictive heart

disease of unknown etiology without pericardial and

endocardial pathology [8,9]. EMF is the commonest

restrictive cardiomyopathy reported in India, especially

from the southern zones. There are very few reports of

RCM from other parts of the country, which do not fall in

the geographic belt of EMF [10–14]. In the present study,

we highlight the clinical and histopathological profile of

patients with RCM from a large tertiary care center in

North India. This is one of the largest series of patients

with RCM unrelated to infiltrative disorders.

by guest on June 5, 2013rdjournals.org/

2. Material and methods

The medical records of all patients with hemodynamic

and angiographic evidence of RCM (i.e. elevated ventric-

ular end diastolic pressures and relatively preserved sys-

tolic function) on cardiac catheterization, between 1985

and 1999 were reviewed. For a diagnosis of restrictive

cardiomyopathy, it was mandatory that the patient had (1)

systemic congestion along with elevated ventricular end-

diastolic pressures; (2) dilated atria; and (3) normal or near

normal ventricular size and function. Ventricular hyper-

trophy was also excluded. All patients also had hemody-

namics consistent with restrictive physiology. Thus the

diagnosis of restrictive cardiomyopathy was initially made

on clinical evaluation along with echo and then subse-

quently confirmed by doing a hemodynamic study. Patients

Table 2

Echocardiographic abnormalities in the two groups

Apex No apex

obliteration obliteration

EMF 20 10

(n= 30)

IRCM 4 18

(n= 22)

who did not have clinically significant heart failure were

excluded. The information obtained from the clinical

examination, electrocardiography, chest X-ray, echocar-

diography, cardiac catheterization and endomyocardial

biopsy (EMB) were recorded. Patients with evidence of

thickened pericardium on echocardiography/computerized

tomographic scan of the thorax or open thoracotomy were

excluded from the study population.

The angiocardiograms of all patients were reviewed by

two authors’ independently and patients were classified

into EMF and non-EMF groups, based upon the criteria

described by Tharakan et al. [14]. The diagnosis of EMF

was made if there was evidence of left or right ventricular

apex obliteration. If there was no apical obliteration, and

the ventricular function was near normal along with

elevated end diastolic ventricular pressures, a diagnosis

of IRCM was made.

Histopathology on EMB of all patients was re-studied in

an effort to identify histological features differentiating

between the two groups as well as to rule out infiltrative

disorders. Changes in the myocardium including fibrosis,

myocytolysis and inflammatory cell infiltration were

graded 0 to 3+; endocardial thickening was also graded

from 0 to 3+.

The investigation conforms with the principles outlined

in the ‘Declaration of Helsinki’ (Br Med J 1964; ii:177).

2.1. Statistical methods

Statistical analysis was performed using X2 test to

compare the two groups and a P value of < 0.05 was taken

as significant.

Tricuspid Mitral Restricative

regurgitation regurgitation doppler flow

9 18 30

3 2 22

Fig. 2. Echocardiogram of patient with IRCM, showing normal sized

ventricles with normal function and enlarged left and right atria.

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729 725

by guest on Junhttp://eurjhf.oxfordjournals.org/

Dow

nloaded from

3. Results

Seventy patients with hemodynamic evidence of

restrictive cardiomyopathy (RCM) were included in the

study. Eighteen had biopsy evidence of cardiac amyloi-

dosis and were excluded from the study. Based upon

angiocardiography, 30 patients who were diagnosed to

have EMF were categorized as Group I. Group II con-

sisted of the remaining 22 patients who did not have

angiographic evidence of EMF and were diagnosed to

have idiopathic restrictive cardiomyopathy according to

the criteria mentioned above. The baseline clinical profile

of the two groups of patients is summarized in Table 1.

Majority of the patients in both groups were young, male

and in NYHA functional class III–IV. There was no

significant difference in the baseline clinical profile

between the two groups.

The ECGs and chest X-rays at presentation are shown

in Table 1. Most of the patients were in sinus rhythm,

and many had evidence of right and left atrial enlarge-

ment. The cardio thoracic ratio was slightly increased in

both groups mainly due to atrial enlargement but as such

there were no distinctive features differentiating the two

groups.

3.1. Echocardiography Findings (Figs. 1, 2)

The diagnostic echocardiographic features of EMF

including cavity obliteration and right ventricular outflow

tract dilatation with varying degrees of atrioventricular

valve regurgitation were present in 20 of 30 patients of

Group I. The other ten patients in group I had evidence

of restrictive heart disease in the form of biatrial enlarge-

ment, dilated inferior vena cava, restrictive doppler flows,

relatively preserved systolic function, and tricuspid valve

Fig. 1. Echocardiogram of patient with EMF, showing apex obliter

regurgitation. In Group II, 4/22 patients showed features

of EMF including cavity obliteration, whereas in the

remaining patients only a restrictive pattern was seen.

Ten patients (four in Group I and six in Group II) were

reported to have mildly reduced LV systolic function

(mean ejection fraction (40F 7%), however, ventricular

cavity size was normal. No patient had evidence of intra-

cavitary thrombus (Table 2) (Fig. 1 and Fig. 2).

3.2. Cardiac Catheterization and angiocardiography

findings

Hemodynamic data from the two groups of patients are

summarized in Table 3. In both the groups, patients had

ation (RV) with dilated right atria and normal left ventricle.

e 5, 2013

Table 3

Hemodynamics and angiography

Hemodynamics (pressures as mmHg) Angiography

RA RVedp PA PAW LVedp Cardiac RV apex LV apex Mild LV TR MR

mean mean index obliteration obliteration dsyfunction

(l/min/m2)

EMF 18.5F 5 20.1F 5.6 33.4F 11.7 23.2F 7.9 22.1F 6.9 2.1F 0.6 30 28 4 21 21

(n= 30)

IRCM 17.2F 4.9 18.1F 4.7 27.8F 7.1 20.9F 6 21.8F 6 2.2F 0.4 0 0 6 0 10

(n= 22)

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729726

by guehttp://eurjhf.oxfordjournals.org/

Dow

nloaded from

markedly elevated ventricular filling pressures with charac-

teristic restrictive hemodynamic pattern. There were no

characteristics differentiating features between the two

groups.

3.3. Angiocardiography (Figs. 3, 4)

Angiocardiography formed the basis of differentiating

the two groups of patients (Table 3). The left and right

ventricle angiograms were unremarkable in the patients

in group II. There was preservation of normal trabecular

pattern and absence of obliterative changes in these

patients. Mild atrioventricular regurgitation was seen in

10 out of the 22 patients. In group I most of the patients

had biventricular involvement, with only two having

isolated right ventricular involvement. The obliterative

changes were more prominent in the right ventricle with

less extensive changes in the left ventricle. Of the 28

patients with left ventricular involvement, 21 had mild to

moderate mitral regurgitation whereas 21/30 patients with

Fig. 3. RV angiogram of a patient with RV EMF showing right ventricular o

right ventricular involvement had significant tricuspid

regurgitation (Fig. 3 and Fig. 4).

3.4. Endomyocardial Biopsy (Figs. 5, 6)

Adequate biopsy tissue was obtained in all patients with

the exception of four patients in Group I.

In Group I (EMF) 17/30 patients (56%) had varying

degrees of interstitial fibrosis, compared to 10/22 patients

(46%) in the IRCM group (Group II) (Table 4). One

patient with EMF had endocardial thickening and infil-

tration with lymphocytic inflammatory cells. Focal infil-

tration with lymphocytes was seen in 11 patients with

EMF. Eosinophilic infiltration was not seen in any

patients. Myocytolysis was seen in 12 (40%) EMF

patients as compared to 11 (50%) IRCM patients. Mild

to moderate myofibre hypertrophy was observed in a

large number of patients (70% EMF and 82% IRCM).

The only major difference between the two groups was

in terms of endocardial thickening, 23 (77%) EMF

utflow tract dilation, RV apex obliteration, and tricuspid regurgitation.

st on June 5, 2013

Fig. 4. RV angiograms of a patient with RV IRCM showing normal

ventricle size and normal trabecular pattern.

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729 727

http://eurjhf.oxforD

ownloaded from

patients had varying degrees of endocardial thickening

while only five (21%) IRCM patients had mild endo-

cardial thickening (Figs. 3 and 4). The endocardial

thickening was predominantly due to cellular fibrocolla-

gen tissue with variable amounts of elastic tissue. There

was evidence of organized thrombus with endocardial

thickening in two patients (Fig. 5 and Fig. 6).

Table 4

Endomyocardial biopsy

Endocardial Myocyte Myocytolysis Interstitial Interstitial

thickening hypertrophy inflammation Fibrosis

EMF 23 (77%) 21 (70%) 12 (40%) 0 17 (56%)

(n= 30)

IRCM 5 (23%) 18 (82%) 11 (50%) 10 (46%)

(n= 22)

by guest on June 5, 2013djournals.org/

4. Discussion

EMF is the commonest form of RCM encountered in

tropical regions including southern parts of India [15–18].

IRCM is a recently described clinical entity and is

characterized by clinical and hemodynamic findings of

restrictive heart disease in the absence of discernible

morphological features [8,9]. These patients have evi-

dence of heart failure with enlarged left and right atria

with both left and right ventricles being normal in size

and function. The Doppler pattern is restrictive with

increased rate of early filling and shortened deceleration

time. The diagnosis of IRCM can be made once all

known causes of restrictive cardiomyopathy have been

excluded. There is very limited endomyocardial biopsy

data available on this condition.

In a recent report from the Mayo clinic, data from 94

patients were presented. These patients were elderly (mean

age 64 years), though the age range was from 10 to 90

years [9]. The majority of the patients were in NYHA

functional class I to II (19% in class I, 53% in class II). A

large number of the patients were in atrial fibrillation

(74%) and 50% died during follow up. Endomyocardial

biopsy data from 33 patients showed interstitial fibrosis

(81%), myocyte hypertrophy (86%), myocyte attenuation

(27%) and degeneration (33%). Endocardial fibrosis was

present in 45% with no inflammatory changes. They had a

typical restrictive filling pattern showing increased mitral E

velocity, increased ratio of mitral early to late filling

(EA > 2) and shortened deceleration time ( < 150 ms).

Hemodynamic data revealed elevated end diastolic pres-

sures though typically the right ventricular end diastolic

pressures were 5 mmHg lower than the left ventricular end

diastolic pressures. The 5-year survival was 64% and 10-

year survival was 37%.

The clinical spectrum of RCM in northern India in our

report seems to have a mixed pattern. There was nearly

equal representation of IRCM and EMF in RCM patients

seen in a large referral hospital.

All of our patients were young and had severe form of

disease as compared to the Mayo clinic study where the

patients were older and less symptomatic. The majority of

our patients were in sinus rhythm unlike the Mayo clinic

study where three-quarters of patients were in atrial fibril-

lation. There were no clinical, electrocardiographic or

radiological features on chest X-ray, which could differ-

entiate between EMF and IRCM. The echocardiographic

feature of ventricular cavity obliteration was a useful

parameter in diagnosis of EMF. However, the cavity

obliteration seen in four IRCM patients, which was absent

on angiocardiography is difficult to explain. As this is a

retrospective study over a decade, variability between

different echocardiographers could partially account for

this.

The IRCM patients with normal ventricular angio-

grams may represent an early stage of the disease con-

tinuum of EMF. However, the hemodynamic data of our

patients (Table 3) clearly highlights the severe stage of

restriction in both the groups. Moreover the age distribu-

tion and symptom duration were similar in both groups.

Thus, the evidence from our observations from a large

number of patients points towards a separate clinical

entity of IRCM.

The degree of endocardial thickening of EMB tissue

was the only histopathologically useful parameter identi-

fying patients with EMF. Seventy seven percent of the

EMF patients had endocardial thickening vs. 21% of

IRCM patients. Interstitial fibrosis was present in both

groups (56% of EMF patients and 46% of IRCM). The

presence of myocytolysis (40% of EMF and 50% of

IRCM) and mild to moderate myofiber hypertrophy

(70% of EMF and 82% of IRCM) was similar in both

groups. Similar kinds of histopathological changes are

Fig. 5. Histopathological specimen from the endomyocardial biopsy of a patient with endomyocardial fibrosis, showing the thick layer of endocardial

thickening, along with myocyte hypertrophy and interstitial fibrosis.

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729728

by guest on June 5, 20http://eurjhf.oxfordjournals.org/

Dow

nloaded from

described in the restrictive cardiomyopathy patients from

the Mayo clinic. Though EMB has no diagnostic value in

patients with restrictive cardiomyopathy, it helps in

excluding infiltrative disorders. Endomyocardial biopsy,

which is a blinded procedure, has its limitations because

tissue may be sampled from an unaffected site [19]. This

may account for the seven EMF patients who had no

endocardial thickening on microscopy.

An interesting feature in the EMF group was that the

pattern of ventricular involvement was different to that

reported from south India. Only two of our patients had

isolated right ventricular EMF and all the rest had signifi-

cant additional involvement of the left ventricle. In a report

by Tharakan et al. [14] of 106 patients, 40 patients had

isolated RV involvement. The etiology of both EMF and

Fig. 6. Histopathological specimen from the endomyocardial biopsy of a

patient with idiopathic restrictive cardiomyopathy, showing the absence of

endocardial thickening (haematoxylin eosin staining) along with presence

of myocyte hypertrophy and interstitial fibrosis.

IRCM are unknown though a geochemical basis for EMF

has been suggested by Valiathan et al. [20].

4.1. Limitation of the study

As a retrospective study extending over a decade the

study is limited by multiple observer variability and con-

trolling the quality of data collection. However, important

data on which the conclusions are based, was collected by

review of electrocardiogram, chest X-ray, tape recordings of

echocardiographic examination, catheterization pressure

trace, angiocardiography films and permanently fixed and

stained tissue slides.

13

5. Conclusions

EMF is not an uncommonly encountered form of RCM

in North India. IRCM is equally prevalent and represents a

separate clinical entity. This report highlights the usefulness

of endomyocardial biopsy and angiography in differentiat-

ing these patients. Endomyocardial biopsy revealed a rela-

tively normal endocardium in IRCM patients. The other

interstitial and myocardial lesions seem to be non-specific.

Endomyocardial biopsy does not contribute to the under-

standing of the pathogenesis of IRCM but rules out any

underlying inflammatory process as none of the patients had

any inflammatory cells in the histopathology. Both EMF

and IRCM have similar clinical features and the patients are

young and quite symptomatic. Echocardiographic evidence

of apex obliteration and endocardial thickening on EMB

help to distinguish EMF from IRCM. These IRCM patients

differ from those seen in temperate climates where the

patients are older and less symptomatic.

S. Seth et al. / The European Journal of Heart Failure 6 (2004) 723–729 729

http:/D

ownloaded from

References

[1] Report of WHO-ISFC task force on definition of cardiomyopathies.

Br Heart J 1980;39:263.

[2] Wood P. Chronic constrictive pericarditis. Am JCardiol 1961;7:48–61.

[3] Roberts WC, Waller BF. Cardiac amyloidosis causing cardiac dys-

function: analysis of 54 necropsy patients. Am J Cardiol 1983;52:

137–46.

[4] Stein E, Stimmel B, Siltzbach LE. Clinical course of cardiac sarcoi-

dosis. Ann NY Acad Sci 1976;278:470–4.

[5] Oslen LJ, Edwards WD, Holmes DR. Endomyocardial biopsy in

hemochromatosis: clinicopathologic correlates in six cases. J Am Coll

Cardiol 1989;13:116–20.

[6] Cherian G, Vijayaraghavan G, Krishnaswami S. Endomyocardial fib-

rosis: report on the hemodynamic data in 29 patients and review of the

results of surgery. Am Heart J 1983;105:659–66.

[7] Metras D, Coulibaly AQ, Quattara K. Recent trends in the surgical

treatment of endomyocardial fibrosis. J Cardiovasc Thorac Surg

1987;28:607–13.

[8] McManus BM, Bren GB, Robertson EA. Haemodynamic cardiac

constriction without anatomic myocardial restriction or pericardial

constriction. Am Heart J 1981;102:134–6.

[9] Ammash NM, Seward JB, Bailey KR, Edwards WD, Tajik AJ. Clin-

ical profile and outcome of idiopathic restrictive cardiomyopathy.

Circulation 2000;101:2490–6.

[10] Valiathan MS, Balakrishnan KG, Kartha CC. A profile of endomyo-

cardial fibrosis. Indian J Pediatr 1987;34:229–36.

[11] Talwar KK, Kumar V, Chopra P, Sharma S, Bahl VK, Srivastava S,

et al. Endomyocardial biopsy study in patients with restrictive car-

diomyopathy. In: Valiathan MS, Somers K, Kartha CC, editors.

Endomyocardial fibrosis. Delhi: Oxford University Press, 1993.

p. 185–94.

[12] Chopra P, Narula J, Talwar KK, Kumar V, Bhatia ML. Histomorpho-

logic characteristics of endomyocardial fibrosis: an endomyocardial

biopsy study. Hum Pathol 1990;21:613–6.

[13] Wahi PL, Eapen K. Clinical profile of endomyocardial fibrosis in

north India. In: Sapru RP, editor. Endomyocardial fibrosis in India.

New Delhi: ICMR, 1983. p. 27–35.

[14] Tharakan JM, Venkitachalam CG, Balakrishnan KG. Angiographic

features of endomyocardial fibrosis. In: Valiathan MS, Somers K,

Kartha CC, editors. Endomyocardial fibrosis. Delhi: Oxford Univer-

sity Press, 1993. p. 168–84.

[15] Beford DE, Konstam GLS. Heart failure of unknown aetiology in

Africans. Br Heart J 1946;8:236–7.

[16] Connon DH, Somres K, Hutt MSR. Endomyocardial fibrosis in

Uganda. Part I: an epidemiologic, clinical and pathologic study.

Am Heart J 1967;74:687–701.

[17] Brockington IF, Edington GM. Adult heart disease in western Nigeria.

Am Heart J 1972;83:27–40.

[18] Nair DN. Endomyocardial fibrosis in Kerala. Indian Heart J

1982;34:412–7.

[19] Mason JW, O’Connell JB. Clinical merit of endomyocardial biopsy.

Circulation 1989;79:971–9.

[20] Valiathan MS, Kartha CC, Eapen JT, Dang HS, Sunta CM. A geo-

chemical basis for endomyocardial fibrosis. Cardiovasc Res July

1989;23(7):647–8.

by guest on June 5, 2013/eurjhf.oxfordjournals.org/