Echocardiographic evaluation of patients with systemic sarcoidosis

Echocardiographic evaluation systemic sarcoidosis

of patients

with

Echocardiographic evaluation of 42 patients with sarcoidosis disclosed 13 patients (group A) with abnormalities compatible with sarcoid heart involvement such as thickening or thinning of the septum (eight patients), pericardial effusion (four patients), and increased end-diastolic dimension of the left ventricle with decreased systolic function (three patients). The remaining 29 patients (group B) were diagnosed as having normal echocardiograms. The clinical data reavealed no statistically significant difference between the groups regarding age, sex, chest x-ray stage, activity, and previous heart disease. Group A patients had older clinical onset of the disease (52 vs 83 months; p < 0.05) and higher incidence of ECG abnormalities than group B patients. There were no statistically significant differences between the groups regarding two-dimensional echocardiographic internal dimensions of both ventricular chambers. The radionuclear right ventricular ejection fraction was low in both groups and the left ventricular ejection fraction was depressed in group A patients (p < 0.01). As observed in pathologic studies, the septum is a target structure which can be characterized echocardiographically. Screening suspected sarcoid heart disease involvement is important to characterize patients with a relatively high risk of clinical cardiac abnormalities such as complete atrioventricular block, ventricular arrhythmias, congestive heart failure, and sudden death. (AM HEART J 110:116, 1985.)

Ruben F. Lewin, M.D., Ram Mor, M.D., Shimon Spitzer, M.D., Alexander Arditti, M.D., Charles Hellman, M.D., and Jacob Agmon, M.D. Petah Tikva and Tel Aviv, Israel

Sarcoid heart involvement is demonstrated by postmortem studies in 20% to 30% of patients with systemic sarcoidosis, l-3 but is rarely established during life. The clinical features of sarcoid heart disease are right and left congestive heart failure, ventricular arrhythmias, conduction disturbances leading to complete atrioventricular block,4*5 pericardial effusion,6 and sudden death, particularly in youngsters.2 Pathologic features of sarcoid heart disease include granulomatous infiltrations of myocardium (interventricular septum, left ventricular free wall, and papillary muscles) and fibrous scar~‘-~; both features are potentially recognizable by echocardiography. The purpose of this study was to assess the frequency and characteristics of echocardiographic abnormalities in a group of biopsy-proven systemic sarcoidosis patients (without coronorariographic or myocardial biopsy procedures) in order to characterFrom the Israel and lone Massada Center for Heart Diseases and the Pulmonary Institute, Beilinson Medical Center; and the Tel Aviv University Sackler School of Medicine. Received accepted

for publication March 1, 1985.

Reprint Diseases,

requests: Beilinson

116

Ruben Medical

Aug.

21, 1984;

revision

received

Jan.

F. Lewin, M.D., Massada Center Center, Petah Tikva 49100, Israel.

29, 1985; for

Heart

ize these patients tory. METHODS Patients.

and to assess their natural

his-

All 42 patients (24 women and 18 men) with clinical signs and symptoms compatible with sarcoidosis followed-up in the Pulmonary Institute (referral center) between 1973 and 1983 were the material of the study. Four other patients were not included becauseof unsuitable echocardiographic examination (three patients) and concomitant severe mitral and aortic valve disease(one patient) of rheumatic origin. The diagnosis of sarcoidosis was established in 19 patients by organ biopsy, in 15 patients by KveimSiltzbach test, and in eight patients by both procedures. Ages ranged between 18 to 76 years (mean 49.4 years). Sarcoidosiswaspresent for lessthan 1 year in six patients (subacute) and from 1 to 15 years in the remaining 36 patients (chronic). At the time of the study, the disease was active in 12 patients. Six patients had no evidence of intrathoracic disease;12 patients manifested stage I (bilateral hilar adenopathy), 13 showedstageII (bilateral hilar adenopathy and pulmonary infiltrates), and 11 manifested stage III (pulmonary infiltrates only). Twenty-three patients were on steroid therapy (10 to 20 mg prednisone daily) at the time of the study. All patients had la-lead ECG, pulmonary function studies,cutaneous’delayedhypersensitivity tests for aner-

Volume Number

110 1, part

1

1. Septal hypertrophy (s, between upper arrows) and normal thickness of the posterior wall (p, between lower arrows) are shown. cm = centimeter; RV = right ventricle; PM = papillary muscle. Fig.

gy (purified protein derivative of tuberculin, Candida sp, Tricophyton sp, and streptokinase-streptodornase),7 quantitative T cell determination, and evaluation of T cell function (graft versus host reaction). Echocardiographic study. M-mode and two-dimensional echocardiographic studies were performed using a Picker C-80 or a Diasonics CV-100 mechanical sector units, in the supine 15- to 30-degreeleft lateral decubitus position. Two-dimensional echocardiographic studies were recorded on video tape for subsequentslow-motion and stop-frame imaging. The video tape images were interpreted by two different observerswithout knowledge of clinical or radionuclear data. The right and left ventricular anatomy, wall thickness, and wall motion were examined in the parasternal (long- and short-axis) and apical (two- [RAO equivalent] and four-chamber) views.8 Right and left ventricular dimensionsand percentage fractional shortening of different axes of both cavities were assessed from the apical four-chamber view according to a modified approach of the method proposed by others.8,9Both ventricular endocardial internal borders were carefully underlined after optimal gain setting. In technically adequate images, the long-axis dimension (right or left ventricle) wasmeasuredfrom the apex to the midpoint of the tricuspid or mitral anulus plane.‘OThe short-axis dimensionwasmeasuredbetween the interventricular septum and the free lateral wall (for the right ventricle) and the posterolateral wall (for the left ventricle) perpendicular to the first third-point of the long-axis dimension. End-diastolic and systolic dimension, and thickness of septum and posterior wall (under the mitral valve leaflets) were measuredat the peak of the R wave in the ECG (at end diastole) and maximal endocardial inward motion was determined (at end systole). The

Echocardiography

in systemic

sarcoidosis

117

Fig. 2. M-mode recording showing asymmetric septal hypertrophy (s, between upper arrows) and normal posterior wall thickness (p, between lower arrows). Normal wall motion and internal dimensionsare shown.

percentage fractional shortening for each axis was calculated. All patients were askedto have at least two echocardiographic examinations to assess reproducibility of intracardiac measurementsover time (2 weeksto 6 months). In no instance was a variation of more than t- 6% in wall thickness or dimensions found. The two-dimensional echocardiography interobserver difference in the assessment of wall thickness and internal dimensions was 4.2 + 0.8% for the right ventricle and 3.8 1 1% for the left ventricle. For each dimension three consecutive sinus beats (or pacemaker beats in casesof complete atrioventricular block) were analyzed and the average was calculated. Radionuclear angiography studies. Radionuclear angiography determinations were done within 24 hours of the echocardiographic study. Labeling of the red blood cells was performed in vivo with stannous pyrophosphate, followed by an injection of 15 mCi of technetiumWm pertechnetate. After readjusting the view in order to obtain the maximal separation between both ventricles (approximately 45-degree left anterior oblique), multigated equilibrium imaging (MUGA) was begun using an Elscint Apex 430 large-field camera and integrated computer systems.The camera was fitted with a very highresolution, low-efficiency APC-4 collimator (Elscint Ltd., Haifa, Israel). Acquisition was performed using a 64 x 64 pixel matrix at 2-powerzoom with a frame rate adjusted to give 24 framesper cardiac cycle. A minimum of 7.5 million total counts per study wasobtained in eachpatient. After finishing acquisition, data analysis were accomplishedby near total automated processingfor both right and left global ventricular ejection fraction. The row data were smoothed once by a nine-point smoothing program; regions of interest were drawn separately around the two ventricles, and a secondderivative edgedetection program

118

Lewin et al.

Table

I. Clinical characteristics

Age (mean + SD) Range Sex (male) Roentgenographic (mean + SD) Activity

American

stage

July, 1995 Heart Journal

Group A (13 pts)

Group B (29 pts)

53 * 10 33-71 8/13 (61% ) 1.5 * 1

48.5 f 17 18-76 10/29 (41% ) 1.9 2 1

NS

4/13

8129 (28%)

NS

8113 2113 2113 4113

2129 9129 0129 0129

<0.005 NS <0.05
o/13 o/13

3129 3129

NS NS

83 f 88 6-180 3/13

52 k 44 2-180 7129

<0.05

(21%)

P

NS NS

ECG abnormalities ST-T changes RAD CAVB RBBB

Other

cardiac

AH AP Onset of symptoms (months) (mean Range Smoking

diseases

+ SD)

Abbreviations: AH = arterial RBBB = right bundle branch

hypertension; block.

AP = angina

pectoris;

CAVB = complete

was applied to define the ventricular edges.Background regions were automatically placed around the two ventricles. Histograms of counts of all the frames of the cardiac cycle for each region of interest were generated; for each ventricle and its associated background, the regions of interest were normalized for area.Background curves were then subtracted from the ventricular region of interest histograms. Normal values. We examined 20 normal physician volunteers by echocardiography and radionuclear angiography. Their mean agewas 31.7years; they had no history of cardiovascular diseases,and their physical examinations, blood pressure, and ECGs were normal. Echocardiographic criteria for diagnosis of suspected sarcoid heart disease. According to anatomopathologic

studies,1~3 involvement of the upper portion of the septum leading to granulomatous infiltration (asymmetric septal hypertrophy) (Figs. 1 and 2) or septal scars (Fig. 3), pericardial effusion or congestive cardiomyopathy were considered suggestive of sarcoid heart disease.When a clinical history of arterial hypertension and symmetric left ventricular hypertrophy waspresent alone (two patients), the echocardiographic study was considered normal. In one patient a small isolated pericardial effusion (
AV block;

NS = not significant;

NS RAD = right

axis deviation;

sarcoid heart disease patients (group A). The other 29 patients were considered as being free of sarcoid heart involvement (Group B). There were no statistically significant differences in age, sex, chest x-ray stage, activity of sarcoidosis, smoking history, incidence of arterial hypertension, or history compatible with concomitant coronary artery disease between the two groups. The time elapsed from onset of symptoms to the performance of the echocardiographic study was longer in those with abnormal echocardiograms (52 + 44 months vs 83 f 88 months; p < 0.05). Electrocardiographic features. The electrocardiogram showed a statistically significant higher incidence of right bundle branch block (p < 0.011, complete atrioventricular .block (p < 0.05) and ST-T changes (p < 0.005) in patients of Group A. Laboratory features (Table II). There were no statistically significant differences in mean vital capacity, mean T lymphocytes percentage, positive graft vs. host reaction, incidence of cutaneous anergy or extrapulmonary sarcoid involvement between the two groups. Echocardiographic

findings.

M-mode (Table 111). The mean septal thickness was similar between the groups; however, there were eight abnormal measurements (four thinning and four thickening) in the abnormal echo group, as stated in the inclusion criteria (group A). On the other hand, there was no difference in the mean posterior wall thickness or number of abnormal posterior wall measurements between the two

Volume

110

Number

1, part

Table

I 19

in systemic sarcoidosis

II. Laboratory features

Vital capacity (“, ) (mean ? SD) Range 7’ cells (mean + SD) R.ange Graft vs host reaction Anergy Steroid treatment Abbreviation:

Echocardiography

1

Group A

Group B

(13 pts)

(29 pts)

P

84 k 14 56-100

36 f 13 54-100

NS

53 * 12 36-80

62 + 10 36-69

NS NS

3/13 10/u 7113

9129 17129 16/29

NS NS NS

‘cm.

RV. .,-

1

.

LV

NS = not significant.

groups (5 of 13 vs 3 of 29). The mean left ventricular end-diastolic dimension and fractional shortening were similar between the two groups, but the mean E-point septal separation was higher in group A patients (8.4 + 7 vs 3.5 + 3.8 mm; p < 0.01). However, the number of abnormal measurements was similar between the two groups (8 of 13 vs 8 of 29; p = NS). The number of patients with abnormal percentage fractional shortening was greater in group A patients (3 of 13 vs none of 29; p < 0.01). The left atrium was enlarged in group A patients (39.1 ? 3.4 vs 34 + 4.87 mm; p < 0.01); however, the number of abnormal measurements (<40mm) was similar between the two groups (8 of 13 patients vs 7 of 29 patients; p = NS). There were four patients with pathologic pericardial effusions in group A (4 of 13 patients vs 1 of 29 patients; p < 0.05). Two-dimensional echocardiography (Table IV). There was no statistically significant difference between the two groups and the control group in the measurement of the right and left ventricular dimensions and percentage fractional shortening of every dimension in the four-chamber view. A consistently but not statistically significant lower percentage fractional shortening of the right and left ventricle in all dimensions measured was found between the control group and both sarcoid groups. Radionuclear angiographic study (Table V). A statistically significant lower right ventricular ejection fraction was seen in both sarcoidosis groups when compared with the control group (p < 0.01). However, no difference was found between both sarcoidosis groups themselves (38.3 + 12 % vs 42 & 8% ; p = NS). The left ventricular ejection fraction of patients with abnormal echocardiograms (group A) was significantly lower (56 + 9% vs 70 * 13 % ; p < 0.01) than in the normal echo and control group.

Fig. 3. M-mode recording showing septal (s) thinning with paradoxical motion. An enlarged right ventricle with normal posterior wall motion is shown. A W = anterior wall.

DISCUSSION

Most of the data concerning sarcoid heart disease are based on anatomopathologic studies, or are based on isolated case reports in patients with arrhythmias, conduction disturbances, or congestive heart failure. The definitive diagnosis of myocardial sarcoidosis in living individuals is made by histologically proven myocardial involvement, requiring open-heart biopsy or closed endomyocardial biopsy. It is known that the “patchy” or focal presentation of myocardial involvement makes the definitive negation of this pathology by biopsy difficult in some circumstances. On the other hand, both methods, by being aggressively invasive, are not suitable for routine implementation in searching for myocardial sarcoidosis. Noninvasive detection of sarcoid heart disease in patients is warranted in order to characterize those at relatively high risk of lifethreatening events. Recently, thallium-201 scans have been proposed as a noninvasive screening method in order to detect sarcoid heart involvement.“-l3 In one study,” three out of five patients with severe cardiac symptomatology had segmental thallium defects of the left ventricle at rest, compatible with an infiltrative disease of the myocardium and confirmed as being sarcoidosis on biopsy in two of them. In another study,12 44 patients with sarcoidosis and no clinical evidence of severe cardiac involvement were studied, disclosing left ventricle defects on thallium scans in 14 of them (32% ); in five of these patients an exercise-induced left ventricular defect was also

120

Lewin et al.

Table

III. Echocardiographic data (M-mode) (Normal

American

Normal Group

values)

echo group A (13 pts)

Normal Group

July, 1995 Heart Journal

echo group B (29 pts)

P

Septal thickness (5-11 mm)

9.1 t 3.5

8.48 + 1.36

NS

Posterior wall thickness (6-10 mm) LVED dimension (<55 mm) Oo F.S. (<25%) E-point septal separation (<5 mm) Left atria1 dimension (<40 mm) Pericardial effusion

8.4 k 2.21

8.37 Z!I 1.51

NS

48.1 z!z 6

46.6 k 6.12

NS

39.5 + 11 a.4 + 7

42.5 k 7.2 3.58 k 3.80


39.1 * 3.4

34 t 4.87


Abbreviations:

Table

“‘0 F.S. = percentage

4113 fractional

shortening;

LVED

NS

l/29 = left ventricular

end-diastolic

<0.05

dimension.

IV. Echocardiographic data (2DE) Control

Group

Group A (13 pts)

(20pts) Right ventricle Diastole (mm) Systole

(mm)

“, FS Left ventricle Diastole (mm) Systole O’aFS

Abbreviations: dimensional

(mm)

Group B (29 pts)

’

-LADd -SADd -LADS -SADs -LAD -SAD

68.3 30.6 48.8 21.4 31.2 30.6

+ -t xk 2 ?z +

8.3 5.4 10 4 9.6 9

62.9 33.9 49 23.8 29.4 29.4

k + * + k f

8.45 5.07 9.97 4.45 10.6 12.3

58.7 29.6 48.3 21.5 19.3 25.1

+ + + f *

-LADd -SADd -LADS -SADs -LAD -SAD

76.9 42.1 59 30.2 23 28.6

f 2 + k 2 k

12 5 10 6 4 6

70.1 45.5 60.4 36.7 16.8 21.0

* + _f k * k

10.6 8.95 9.2 10.5 10.5 10.6

69.9 43.3 57.9 32.3 19.3 25.1

+ k t k ? +

LAD

= long-axis

?:a FS = percentage echocardiography.

fractional

shortening;

observed (compatible with coronary artery disease). However, two of these patients had normal coronary anatomy. When the abnormal thallium scan group was compared with the normal thallium group, an abnormal higher E point-septal separation measurement was found in the first group. The authors concluded that abnormal thallium scans can distinguish patients with myocardial sarcoidosis, if other diseases which can produce similar results are excluded (coronary artery disease, congestive or hypertrophic cardiomyopathy, hypoxemia, etc.).*2-‘4 In addition to its high cost, and due to a relatively low energy primary photo peak and low targetto-background ratio, imaging with this radiotracer can be difficult.15 Despite the established value of thallium-201 myocardial imaging for detection of fixed obstruc-

dimension;

SAD

= short-axis

dimension;

d = diastole;

s = systole;

12.2 6.8 9.16 8.3 8.2

+ 10

2DE

9 6.5 10.5 7.6 8.4 9.4 = two-

tive coronary artery disease, no studies concerning sensitivity and specificity for detecting biopsy-proven sarcoid heart disease were done. Echocardiographic evaluation of sarcoid heart disease. Echocardiography is a useful noninvasive tech-

nique for screening cardiac anatomy and function for possible markers of sarcoid heart disease such as septal thickness or scars, pericardial effusions, and decreased ventricular systolic function.16 Surprisingly, we did not find any case of aneurysm (except for the septum) or mitral anulus calcification, as expected’ in accordance with previous studies.12 Echocardiographic detection of congestive cardiomyopathy’l and aneurysm of the left ventriclelap lg have been reported. In our material, several echocardiographic markers were found useful, i.e., myocardial thickening, possibly by sarcoid granulomas, or thinning, as a

Volume Number

110 1. part

Echocardiography

1

late result of the same process. Since no coronary angiography or myocardial biopsy was obtained in the patients, the possibility of concomitant coronary heart disease and myocardial sarcoidosis cannot be excluded. Markers of systolic function of the left ventricle were within normal limits (percent fractional shortening and left ventricular end-diastolic dimension), as expected from previous studies; however, in three patients end-diastolic dimension and E-point septal separation distance were out of normal limits (x2 = p < 0.01). It is reasonable to assume that sarcoid heart involvement (granulomatous infiltration) is initially seen as a mild thickening of myocardium (principally interventricular septum), leading to asymmetric septal hypertrophy in some cases. Later, as the result of the natural history of the disease or corticosteroid treatment,2 resolution or scar occurs, as in the lungs.2o It is in this latter stage when aneurysm and/or congestive cardiomyopathy is diagnosed. It was shown in previous studies’s’ that the septum is compromised in 90% of autopsy cases with cardiac sarcoidosis, and the posterior part of the septum appears to be more commonly involved.3 In accordance with these observations, echocardiography appears to be a suitable noninvasive technique to assess this sarcoid target structure (8 of 13 patients with abnormal echocardiogram had half septal thickening and half thinning). Pericardial effusion appears to occur in 10% to 15 % of living individual@ with systemic sarcoidosis, a percentage similar to our series. Its etiology is not clear; however, it has been stated2 that 3% to 5% of patients dying from sarcoidosis have this pathology as the leading cause of death. Correlation

between

ECG and echocardiography.

It

has been well known that patients with myocardial sarcoidosis have an increased incidence of ECG abnormaIities4* 5 when compared with myocardial sarcoid-free individuals and sex-matched healthy controls.21 Conduction disturbances, affecting principally the septal portion of the system, appeared to be a marker for heart involvement (bundle branch block; AV nodal involvement leading to first, second, and third-degree AV block; and infarct-like septal Q waves).21 In our material, a higher incidence of ECG abnormalities was indeed found in echoabnormal patients, and a strong correlation between septal distorted anatomy and ECG findings was uncovered. It is possible that this apparently early marker can be used as an indication of sarcoid heart disease. Corticosteroid treatment. No significant statistical

Table

V. Radionuclear

Parameter and normal ualues RVEF (“0) (54.3 It 6.6) LVEF ( “;t ) (63.7 k 6.4)

in

angiography

systemic

strrcoidosis

121

data

Group A (13 pts)

P

38.3 t 12.6

NS

56 t 9


Abbreviations: LVEF = left ventricular ventricular ejection fraction.

ejection

fraction;

RVEF

= right

difference was found between the two groups regarding the time and dose of corticosteroid administration. However, a prospective study is warranted in order to make clear-cut conclusions. Despite the conflicting evidence regarding the use of corticosteroids in the treatment of pulmonary sarcoidosis which has been presented,22 the trend is to treat these patients on long-term course~.~~ Similar trends have to be adopted when considering sarcoid heart disease treatment. In some cases, cardiac abnormalities are the presenting symptoms of systemic sarcoidosis, and corticosteroids are the only treatment available in cardiomyopathy.16 Cor pulmonale and sarcoidosis. Consistent decreased echocardiographic and radionuclear right heart function, without chamber enlargement, was seen in both groups, apparently as a common response to pulmonary sarcoid involvement. Therefore it seems that abnormalities of the right side of the heart are difficult to evaluate when sarcoid heart disease is suspected. Conclusions. Echocardiography is a useful noninvasive technique for the characterization of suspected sarcoid-related heart abnormalities. Abnormal septal thickening and thinning, pericardial effusion and, in some cases, decreased systolic function appear to be important markers of this group. In the absence of obvious pulmonary or extrapulmonary involvement, the diagnosis of sarcoidosis may be overlooked in a situation of potentially treatable cardiac disease. Echocardiography offers a cost-effective, low-risk technique to detect possible sarcoid heart disease. REFERENCES

1. Silverman KJ, Hutchins GM, Buckley BH: Cardiac sarcoidosis: A clinico-pathologic study of 84 unselected patients with systemic sarcoidosis. Circulation 58:1204, 1978. 2. Roberts WC, McAllister HA Jr, Ferrans V,J: Sarcoidosis of the heart. A clinicopathologic study of 35 necropsy patients (Group I) and review of 78 previously described necropsy patients (Group II). Am J Med 83:68, 1977. 3. Matsui Y, Iwai K, Tachibana T, Fruie T, Shigematsu N, Izumi T. Homma AH, Mikami R, Hongo 0, Hiraga Y,

122

4. 5. 6. 7. 8. 9.

10. 11. 12.

13.

Lewin et al. Yamamoto M: Clinicopathological study on fatal myocardial sarcoidosis. Ann NY Acad Sci 278:455, 1976. Goodwin JF, Roberts WC, Wenger NK: Cardiomyopathy. In Hurst JW editor: The heart. New York, 1982, McGraw-Hill Book Co., Inc, p. 1330. Wynne J, Braunwald E: The cardiomyopathies and myocarditides. In Braunwald E, editor: Heart diseases. Philadephia, 1980, W.B. Saunders Co, p. 1464. Kinnev E. Murnhv R. Ascunce G. Donohoe R, Zelis R: Pericardial effusions in’sarcoidosis. Chest 76:476,I979. Topilsky M, Shohat B, Spitzer S, Joshua H: Relationship between T-rosette forming cells and cellular immunity in sarcoidosis. Ann NY Acad Sci 278:108, 1976. Weyman AE: Cross sectional echocardiography. Philadelphia, 1982, Lea & Febiger, Appendix, p 497. Bommer W, Weinert L, Neumann A, Neef J, Mason DT, De Maria A: Determination of right atria1 and right ventricular size by two-dimensional echocardiography. Circulation 60:91, 1979. Schnitt GER I, Gordon EP, Fitzgerald P, Popp RL: Standardized intracardiac measurements of two dimensional echocardiography. J Am Co11 Cardiol 2:934, 1983. Bulkley BH, Roleau JR, Whitaker JQ, Strauss HW, Pitt B: The use of 201 thallium for myocardial perfusion imaging in sarcoid heart disease. Chest 72:27, 1977. Kinney EL, Jackson GL, Reeves W, Zelis R, Beers E: Thallium myocardial defects and echocardiographic abnormalities in patients with sarcoidosis without clinical cardiac dysfunction. An analysis of 44 patients. Am J Med 68:497, 1980. Schafer J, Runge M, Mathey D, Hanrath P, Montz R: Cardiac

American

14.

15. 16. 17. 18. 19.

20. 21. 22. 23.

July, 1985 Heart Journal

involvement in sarcoidosis (abstr). Eur Heart J (Suppl A):63, 1981. Mehrotra PP, Weaver YJ, Higginbotham EA: Myocardial perfusion defect on thallium 201 imaging in patients with chronic obstructive pulmonary disease. J Am Co11 Cardiol 2:233. 1983. Berger HJ, Zaret BL: Nuclear cardiology. N Engl J Med 305:799, 1981. Johnson RA, Palacios I: Dilated cardiomyopathies of the adult (second of two parts). N Engl J Med 307:1119, 1982. Lore11 B, Alderman EL, Mason JW: Cardiac sarcoidosis: Diagnosis with endomyocardial biopsy and treatment with corticosteroids. Am J Cardiol 42:143, 1978. Cepin D, McDonough M, James F: Cardiac sarcoidosis-a case with unusual manifestations. Arch Intern Med 143:142, 1983. Lull RJ, Dunn BE, Gregoratos G, Cox WA, Fisher GW: Ventricular aneurysm due to cardiac sarcoidosis with surgical cure of refractory ventricular tachycardia. Am J Caidiol 30:282, 1972. Stilzbach LE: Sarcoidosis. In Fishman AP, editor: Pulmonary diseases and disorders. New York, 1980, McGraw-Hill Book Co, Inc, p. 905. Thunell M, Bjerle P, Stjenberg N: ECG abnormalities in patients with sarcoidosis. Acta Med Stand 213:115, 1983. Harkleroad LE, Young RL, Savage PJ, Jenkins DW, Lordon RE: Pulmonary sarcoidosis. Long-term follow-up of the effects of steroid therapy. Chest 82:84, 1982. De Remee RA: Sarcoidosis and steroids. Chest 83:289, 1983.