deterioration of cardiac function during the progression

1595

□ REVIEW ARTICLE □

Deterioration of Cardiac Function during the Progression ofCardiac Sarcoidosis: Diagnosis and Treatment

Fumio Terasaki and Nobukazu Ishizaka

Abstract

The cardiac involvement of sarcoidosis causes progressive heart failure symptoms and is a life-threatening

condition; thus, an early and appropriate diagnosis of this condition is crucial. On the other hand, the decline

in the cardiac function is rapid; therefore, patients usually have moderate-severe left ventricular dysfunction

when diagnosed with cardiac sarcoidosis, which may decrease the effectiveness of therapies. We herein report

three illustrative cases of heart failure due to cardiac sarcoidosis in patients who were or were not diagnosed

with preceding systemic sarcoidosis. We also discuss the currently available diagnostic modalities and possi-

ble biomarkers for the diagnosis of cardiac sarcoidosis.

Key words: cardiac sarcoidosis, cardiac function, congestive heart failure

(Intern Med 53: 1595-1605, 2014)(DOI: 10.2169/internalmedicine.53.2784)

Introduction

The clinical presentation of sarcoidosis with cardiac in-

volvement may depend on the extent and location of the

granulomatous processes (1, 2). The most frequent cardiac

manifestations of cardiac sarcoidosis (CS) include conduc-

tion disturbances and arrhythmias, sudden death and conges-

tive heart failure (CHF), the latter being reported in up to

30% of affected patients (3-9). Increased awareness, early

medical treatment, the use of pacemakers and implantable

cardioverter-defibrillators (ICDs) and cardiac resynchroniza-

tion therapy with ICD (CRTD) are changing the causes of

death in CS patients, with CHF becoming the most common

cause.

Yazaki et al. followed 95 CS patients for a mean period

of 68 months. Of these, 40 died, and 73% of the deaths

were associated with heart failure. The hearts of patients

with sarcoidosis-associated CHF often display a similar

morphology to that of patients with idiopathic dilated car-

diomyopathy (DCM). In a study of more than 100 patients

undergoing left ventriculoplasty for idiopathic DCM, we

found that 7% actually had CS, which had been undiag-

nosed prior to the cardiac symptoms (9). Yazaki et al. dem-

onstrated a five-year survival rate of 37% in patients with

sarcoidosis and 64% in patients with idiopathic DCM; both

groups had a similar New York Heart Association (NYHA)

functional class and left ventricular ejection fraction

(LVEF) (10). Therefore, the prognosis of CS may be worse

than that of idiopathic cardiomyopathy, and the onset and

severity of CHF may be a prognostic indicator of CS. On

the other hand, the prognosis of the patients with CS and

DCM has recently been improving due to the use of

evidence-based therapies and/or early detection of the dis-

ease. In a systematic review of the mortality data in patients

with CS, it was found that the five-year survival now ranges

from 75% to 100%, and that the extent of left ventricular

(LV) dysfunction is the strongest predictor of survival (11).

Considering that CS, when untreated, causes progressive

heart failure, which is frequently life-threatening, early de-

tection of the cardiac sarcoid involvement and the initiation

of appropriate therapies is mandatory; however, patients di-

agnosed with CS usually have advanced LV dysfunction at

the time of diagnosis of the CS (12). This is in part due to

the fact that cardiac dysfunction may occur suddenly and

progresses rapidly in the patients with systemic sarcoidosis,

who may not be followed-up very frequently as outpatients.

In the present study, we report three illustrative cases of

CHF and review the clinical issues associated with CS, with

special reference to the deterioration of the cardiac function

Department of Cardiology, Osaka Medical College, Japan

Received for publication February 27, 2014; Accepted for publication March 18, 2014

Correspondence to Dr. Fumio Terasaki, [email protected]

Intern Med 53: 1595-1605, 2014 DOI: 10.2169/internalmedicine.53.2784

1596

Figure　1.　The temporal changes in the LV end-diastolic dimension (LVDd; vertical axis in A, mm) and LVEF (vertical axis in B, %) on the echocardiograms of the three patients. The horizontal axis represents years, and the 0 indicates the time at which cardiac sarcoid involvement was diagnosed. The pink dotted lines represent 55 mm LVDd in A and 50% LVEF in B, which are critical levels considered to indicate the progressive deterioration of the LV function in cardiac sarcoidosis.

10203040506070

-4 -3 -2 -1 0 1 2 3 4

Case 1

Case 2Case 3

Case 1

Case 2

Case 3

A B

Years Years

during progressive congestive cardiomyopathy. We also dis-

cuss the rate of the decrease in cardiac contractility in pa-

tients diagnosed or undiagnosed with systemic sarcoidosis

without known cardiac abnormalities prior to the emergence

of the cardiac manifestations.

Case Presentations

Case 1

A 51-year-old woman was admitted to our hospital with

sudden-onset symptoms of heart failure, which had started

about one month earlier. She reported no history of systemic

sarcoidosis. Echocardiography showed LV enlargement and

a diffusely decreased LVEF. She had undergone echocar-

diography 2.4 years prior, which showed a normal ejection

fraction (Fig. 1). Chest X-rays demonstrated cardiomegaly

and right hilar lymph node enlargement (Fig. 2A). An elec-

trocardiogram (ECG) showed normal sinus rhythm, albeit

with prolonged PQ intervals, incomplete right bundle branch

blocks, low voltages in the limb leads and ST-T abnormali-

ties (Fig. 2B). These changes in X-rays and ECG indicators

were not observed in the ECG obtained 2.4 years earlier, ex-

cept for slight PQ prolongation (Fig. 2C).

On admission, the patient’s plasma brain natriuretic pep-

tide (BNP) levels were markedly elevated (706 pg/mL).

Coronary angiography showed normal coronary arteries, and

left ventriculography showed hypokinesis of the anterior and

inferior walls (Fig. 2D, E). A histological examination of bi-

opsied endomyocardial samples revealed the infiltration of

mononuclear cells and enhanced fibrosis, although

sarcoidosis-like granulomatous formations were not detected

(Fig. 2F). The serum angiotensin-converting enzyme (ACE)

activity (17.0 IU/L) and soluble interleukin-2 receptor (sIL-2

R; 395 U/mL) levels were within the normal limits. There-

fore, we tested her for possible CS in further imaging ex-

aminations.

Cardiac magnetic resonance (CMR) imaging demonstrated

late gadolinium (Gd) enhancement (LGE) from the mid-wall

to the epicardial sites of the anteroseptal, anterolateral and

inferior walls of the left ventricle (Fig. 3A, D). T2-weighted

CMR showed high-intensity signal areas between the endo-

myocardial and mid-ventricular regions (Fig. 3B, E),

whereas 99m-technetium myocardial scintigraphy showed

perfusion defects on the basal areas and on the anterolateral

wall of the left ventricle (Fig. 3C, F). Subsequent 18F-

fluorodeoxyglucose positron emission tomography (18F-FDG

PET) showed increased nuclear uptake in the myocardium

and the hilar, para-aortic and subclavian lymph nodes

(Fig. 4A-C).

The cytology of fine-needle aspiration samples obtained

from the right supraclavicular lymph node clearly showed

multinucleated giant cells (Fig. 4D). To confirm the absence

of tuberculosis, nodal biopsies were examined, which re-

vealed noncaseating-epithelioid cell granulomas with multi-

nucleated giant cells, confirming the diagnosis of sarcoidosis

(Fig. 4E) (13). Following the diagnosis of systemic sarcoi-

dosis with cardiac involvement, prednisolone treatment was

initiated at a dose of 60 mg/day every other day; this dose

was gradually tapered to 20 mg/day every other day and has

been maintained.

Case 2

A 57-year-old woman with lung and lymph node sarcoi-

dosis that had been histologically diagnosed from biopsy

specimens two years earlier was transported by ambulance

to an emergency center after developing cardiogenic shock.

Chest X-rays demonstrated cardiomegaly and pulmonary

congestion. ECG analyses showed ventricular tachycardia

(VT; Fig. 5A), which was cured to sinus rhythm using elec-

trical cardioversion. Coronary angiography showed no sig-

nificant luminal stenosis, but left ventriculography showed

severe diffuse hypokinesis, particularly in the basal regions

(areas 1 and 5). The patient was transferred to our hospital


1597

Figure　2.　The chest X-ray, electrocardiogram (ECG), left ventriculography and cardiac biopsy findings for case 1. The chest X-ray showed cardiomegaly (cardiothoracic ratio of 60%) and enlarge-ment of the right hilar lymph nodes (A). The ECG showed normal sinus rhythms, PQ prolongation, incomplete right bundle branch block, low voltage in the limb leads and ST-T abnormalities (B). An ECG obtained 2.4 years earlier showed only slight PQ prolongation (C). Left ventriculography showed diffuse hypokinesis, particularly in the anterior (areas 1-2, arrows) and inferior walls (areas 3-4, arrow heads; D, end-diastolic phase; E, end-systolic phase). The histological evaluation of the endomyocardial biopsies showed the infiltration of mononuclear cells and enhanced fibrosis (F, scale bar indicates 50 μm).

A B C

D E F

the following day for further examination and treatment.

On admission, a chest X-ray demonstrated cardiomegaly

and an ECG showed normal sinus rhythm, with slight PQ

prolongation, complete right bundle branch block and occa-

sional premature ventricular contraction (Fig. 5C). Echocar-

diography showed LV enlargement and a diffusely decreased

LVEF (Fig. 5B), and the plasma BNP levels were elevated

(181 pg/mL). The echocardiographic data were almost nor-

mal, except for local hypokinesis at the anteroseptal LV wall

of the basal region, which had also been identified during

the examination performed two years earlier (Fig. 1, 5D).

The serum ACE activity (17.0 IU/L) was normal, but the

sIL-2R (771 U/mL) levels were elevated.

CMR imaging showed LGE on the epicardial site of the

left ventricle, but not on the lateral wall (Fig. 6A, B). T2-

weighted images showed hyperintense signals on the endo-

cardial and mid-wall sites of the anterolateral wall and inter-

ventricular septum (Fig. 6C), while 99m-technetium myocar-

dial scintigraphy showed perfusion defects in the basal parts

of the anteroseptal-inferoposterior wall of the left ventricle

(Fig. 6D). The previous 18F-FDG PET analyses had demon-

strated enhanced nuclear uptake in the lung and hilar and

mediastinal lymph nodes, but not in the heart. Cardiac biop-

sies were not obtained, because informed consent was not

obtained. An ICD was implanted and corticosteroid therapy

was initiated at a dose of 60 mg/day every other day; this

dose was gradually tapered to 5 mg/day every day and has

been maintained.

Case 3

In a previous study, we analyzed the clinical history of 54

patients diagnosed with systemic sarcoidosis (14). Among

these patients, two were not diagnosed with CS and were

eventually diagnosed with cardiac involvement during the

follow-up. One of these patients was a 69-year-old man with

systemic, but not cardiac, sarcoidosis confirmed by lymph

node biopsy. The patient had previously undergone periodic

follow-up with electrocardiography and echocardiography

for possible cardiac involvement. Although his LVEF was

within the normal range at the time of follow-up, it sud-

denly declined (Fig. 1). The 18F-FDG PET study showed in-

creased nuclear uptake in the left ventricle (Fig. 7A) and

CMR showed LGE, which was more prominent in the

epicardial regions (Fig. 7B). An endomyocardial biopsy


1598

Figure　3.　The cardiac magnetic resonance (CMR) imaging and 99m-technetium myocardial scin-tigraphy findings for case 1. Late gadolinium enhancement (LGE) was observed at the mid-wall-epi-cardial sites of the anteroseptal, anterolateral and inferior walls of the left ventricle (A and D, white arrows). The short axis shows the black-blood T2-weighted images demonstrating hyperintense sig-nals on the endocardial-mid-wall site of the anterolateral and inferior walls (B and E, yellow arrows). 99m-technetium myocardial scintigraphy showed perfusion defects in the basal part of the anterosep-tal-inferoseptal and the anterolateral wall of the left ventricle (C and F).

A

D

CB

E F

showed fibrotic degeneration with the infiltration of CD45-

positive lymphocytes and CD68-positive monocytes/macro-

phages, although granulomatous changes were not detected

(Fig. 7C). Sustained VT was documented, and after the di-

agnosis of cardiac sarcoid involvement, corticosteroid ther-

apy was initiated and a CRTD was implanted.

As indicated in Fig. 1, the CHF progressed rapidly within

only a few months or years in all three cases, sometimes in

association with ventricular tachyarrhythmias. The data from

the present cases suggest that a LV end-diastolic dimension

<55 mm and a LVEF <50% may be critical factors indicat-

ing the progressive deterioration of the LV function in CS

patients.

Strategies for Diagnosing Cardiac Sarcoidosis

CMR imaging

In cases with cardiac involvement of sarcoidosis, the

myocardium is replaced by fibrotic fibrogranulomatous tis-

sue. These fibrotic changes in the heart can be observed us-

ing magnetic resonance imaging (MRI) with the LGE tech-

nique. Previous studies demonstrated that characteristic LGE

patterns and locations are diagnostic of CS, and that CMR

imaging with LGE facilitates the prediction of the LV func-

tion. T2-weighted MRI has also been used to detect acute

inflammatory processes associated with myocardial sarcoido-

sis in CS patients (15). Therefore, T2-weighted sequences

may provide increased sensitivity for predicting progressive

deterioration in the cardiac function in CS patients.

In case 1 in the present study, CMR imaging demon-

strated LGE that may represent chronic myocardial dam-

age (16) from the mid-wall to the epicardial sites. In addi-

tion, T2-weighted CMR, which can demonstrate acute in-

flammation or edema (16), showed high-intensity signal ar-

eas between the endomyocardial and mid-ventricular re-

gions. These findings collectively suggested that the

inflammation-associated injury had migrated from the

epicardial region to the endomyocardial region.

Matoh et al. studied 12 sarcoidosis patients using Gd-

MRI, myocardial perfusion single photon-emission com-

puted tomography (SPECT) (17), gallium-67 citrate (Ga-67)

scintigraphy and/or 18F-FDG PET. LGE was observed in five

patients, and was positive in 39 (39%) of 100 LV segments.

LGE was distributed primarily from the mid- to the epimyo-

cardium, and the lack of perfusion defects on myocardial

perfusion SPECT was more prominent in less transmural

LGE segments. Two patients with diffuse LGE and one case

with focal LGE exhibited positive cardiac uptake on Ga-67

scintigraphy, while two other patients with focal LGE

showed cardiac uptake on 18F-FDG PET. The authors con-

cluded that LGE was useful for diagnosing CS and evaluat-

ing the cardiac function, suggesting that the distribution of

LGE from the mid- to the epi-myocardium is characteristic


1599

Figure　4.　The 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) and cytohisto-logical findings of the lymph nodes from case 1. The frontal plane FDG PET/CT fusion images dem-onstrated abnormal uptake in the cervical and mediastinal lymph nodes and the anterior and inferior walls (A), horizontal plane, the interventricular septum and the lateral wall (B) of the left ventricle (arrows). The whole-body FDG PET maximum intensity projection image showed abnormal uptake into the general lymph nodes, including the right cervical lymph nodes, from which fine-needle aspi-ration cytology and biopsy specimens were obtained (C). The fine-needle aspiration cytology of the right supraclavicular lymph nodes; Papanicolaou staining of the aspirated specimens demonstrated the three-dimensional appearance of multinucleated giant cells (D). The staining of paraffin-embed-ded lymph node specimens from biopsies revealed noncaseating epithelioid cell granulomas with mul-tinucleated giant cells containing asteroid bodies, confirming the diagnosis of sarcoidosis (E). The scale bars in D and E indicate 50 μm.

A C D

B E

Figure　5.　The ECG and echocardiogram findings for case 2. Ventricular tachycardia was docu-mented at an emergency center (A). Echocardiography performed at this admission showed LV en-largement (LV end-diastolic dimension, 59 mm) and a diffusely decreased LV ejection fraction of 38% (B). An ECG obtained on admission showed normal sinus rhythm, slight PQ prolongation and right bundle branch block (C). The echocardiographic findings obtained two years earlier were al-most normal, except that local hypokinesis in the anteroseptal LV wall of the basal region was ob-served (D).

A CB D

of CS, and that larger LGE areas correlate with poor LV

function.

Ichinose et al. analyzed the topographic localization of

myocardial lesions on CMR imaging and their relationships


1600

Figure　6.　The CMR and myocardial scintigraphy findings for case 2. Long- (A) and short- (B) axis CMR images showed LGE at the epicardial site of the mid-basal portions of the left ventricle (ar-rows), but not the lateral wall. The short axis showed black-blood T2-weighted images demonstrating hyperintense signals at the endocardial and mid-wall sites of the anterolateral-inferoseptal wall (C, arrows). 99m-technetium myocardial scintigraphy showed perfusion defects in the basal part of the anteroseptal-inferoposterior wall of the left ventricle (D).

A CB D

Figure　7.　The 18F-FDG PET, CMR and cardiac biopsy findings for case 3. 18F-FDG PET showed abnormal uptake in the heart (arrows) and cervical, mediastinal and hilar lymph nodes (A). CMR showed LGE at the epicardial site of the anteroseptal-lateral wall of the left ventricle (B, arrows). The histological study of an endomyocardial biopsy sample showed fibrotic degeneration with the infiltra-tion of mononuclear cells (C, scale bar indicates 50 μm). Most mononuclear cells were CD45-positive lymphocytes or CD68-positive monocytes/macrophages (data not shown), although granulomatous changes were not detected.

A CB

with the plasma BNP levels and cardiac function parameters

in 10 CS patients. Myocardial hyperenhancement was sig-

nificantly more common in the subepicardial layers than in

the subendocardial layers (18). Moreover, the authors

showed a significant correlation between the global extent of

LGE and the plasma BNP levels, and a negative correlation

between the LV end-diastolic volume indices and LVEF in

CS patients. Therefore, the extent of myocardial lesions may

be related to the LV dysfunction and plasma BNP levels in

CS patients.

Watanabe et al. retrospectively evaluated 17 CS patients

who were diagnosed according to the 2006 revised guide-

lines of the Japanese Ministry of Health and Welfare (13)

and underwent CMR imaging (19). In that study, the LGEs

were located, and the relationship between the LGE and

LVEF characteristics and duration of sarcoidosis was evalu-

ated. While the LGE was most frequently found in the

subepicardial layer, the affected LGE segments were corre-

lated with the LVEF (r=-0.84, p<0.0001) and LV diastolic

volumes (r=0.88, p<0.0001). Transmural lesions were also

significantly more common in patients with an LVEF of

35% or lower than in those with an LVEF greater than 35%

(p=0.0004). All patients with an LVEF of 35% or lower had

both subepicardial and transmural lesions, and the affected

LGE segments were positively correlated with the duration

of sarcoidosis originating in the extracardiac organs (r=0.76,

p=0.005). The authors concluded that CMR imaging with

LGE facilitates the diagnosis of CS and predicts the LV

function.

Patel et al. performed an observational study of 152 pa-

tients with biopsy-identified extra-cardiac sarcoidosis (20),

no known cardiac sarcoidosis, and an LVEF of �50%. The

presence of LGE in the LV myocardium was considered to

be diagnostic for CS. Compared with patients without LGE,

those with LGE had higher heart rates (84±19 vs. 76±18

bpm, p=0.002), a greater prevalence of abnormal ECGs (76

vs. 31%, p<0.001), more diastolic dysfunction (67 vs. 33%,

p=0.05), a decreased right ventricular ejection fraction (49±8


1601

vs. 55±6%, p=0.012), and evidence of non-sustained VT (33

vs. 6%). The authors concluded that patients with sarcoido-

sis and preserved systolic function commonly had myocar-

dial damage, which may increase the risk of ventricular

tachyarrhythmias.

18F-FDG PET

More recently, 18F-FDG PET has been increasingly used

for the evaluation and monitoring of CS patients (21-23).

Accordingly, 18F-FDG PET imaging plays a significant role

in the clinical diagnosis, assessment of disease activity,

monitoring of therapeutic responses and evaluation of the

prognoses of CS patients (24). In 18F-FDG PET imaging for

CS, the patterns of glucose metabolism and myocardial per-

fusion are known markers of inflammation. Whereas en-

hanced glucose metabolism with normal perfusion indicates

active inflammation, decreased perfusion with high glucose

metabolism represents advanced stage disease, and absent or

decreased perfusion with limited glucose metabolism indi-

cates end-stage CS. It should be noted that, for appropriate

interpretation of the PET images, the suppression of physi-

ological myocardial 18F-FDG PET uptake is crucial, which

can be achieved by fasting, dietary carbohydrate restriction

and the administration of heparin before the nuclear injec-

tion (25).

McArdle et al. (26) compared the locations and degrees

of FDG uptake in 20 CS patients with either advanced atrio-

ventricular block (AVB) or VT and compared the results

with those of seven CS patients without AVB or VT. Both

the mean LV standardized uptake (SUV) values and maxi-

mum SUV in the CS patients with VT were significantly

higher than those in CS patients with AVB (mean SUV: me-

dian VT, 5.33; range 4.7-9.35 vs. median AVB, 2.48; range

0.86-8.59; p=0.016; Max SUV: median VT, 11.07; range,

9.24-14.4 vs. median AVB, 5.63; range 3.42-15.71, p=0.005)

and those in control patients. The SUV values did not differ

significantly between the AVB patients and controls. The

receiver-operator characteristic (ROC) analyses were used to

identify patients with VT, and showed areas under the curve

(AUCs) of 0.93 and 0.895 for a mean LV SUV value of

>3.42 and a maximum SUV value of >8.56, respectively (p<

0.001). Moreover, the mean overall LV SUV value was cor-

related with the number of abnormal segments determined

by a visual analysis (Spearman’s coefficient =0.506, p=

0.006) and was negatively correlated with the resting LVEF

(Spearman’s coefficient =-0.42, p=0.024). In addition, the

mean EF was measured using gated rest PET perfusion

scanning and was found to be lower in VT patients than in

AVB patients (median VT, 33%; range 15-56 vs. median

AVB, 51%; range 18-71, p=0.082) and significantly lower in

VT patients than in clinically silent CS patients (p=0.0026).

These observations suggest that there is a relationship be-

tween the degree of abnormal FDG uptake and the clinical

presentation, particularly with reference to ventricular tachy-

arrhythmia and decreased LV function. Further prospective

studies are required to confirm these relationships.

Mehta et al. (27) evaluated 62 ambulatory patients with

sarcoidosis and diagnosed CS by assessing abnormalities de-

tected using CMR imaging or 18F-FDG PET. CS patients

were referred for risk stratification by electrophysiology

(EPS). Among the 62 patients examined, the prevalence of

CS was 39%, and CS patients were more likely to have ab-

normal Holter ECG recordings (50 vs. 3%, p<0.001) and

echocardiographic parameters (25 vs. 5%, p=0.02). The de-

gree of pulmonary impairment was not predictive of CS.

Two of the 17 patients analyzed using EPS had abnormal

test findings and received ICDs. Over a mean follow-up of

1.8 years, no patient died or exhibited ventricular arrhyth-

mias, and the authors concluded that the sarcoid lesions

identified on 18F-FDG PET are not predictive of arrhythmias

in patients with preserved cardiac function.

Matthews et al. (28) described a patient with CHF and

biopsy-diagnosed sarcoidosis whose imaging data were con-

sistent with cardiac involvement, as evidenced by extensive18F-FDG uptake into the ventricles and multiple focal areas

of myocardial hyperintense signals on T2-weighted CMR

images and corresponding areas of LGE. Significant clinical

improvements were observed after the initiation of medical

therapy for CHF and corticosteroids. Subsequent follow-up

CMR imaging studies performed three and nine months later

demonstrated improvements in the LVEF and the almost

complete disappearance of abnormal T2 signals, with persis-

tent LGE. However, the three- and 12-month 18F-FDG PET

analyses demonstrated complete resolution of all cardiac and

extracardiac sarcoid activity. Therefore, although both the

CMR and 18F-FDG PET analyses facilitated the diagnosis of

CS, only 18F-FDG PET imaging demonstrated serial assess-

ments that reflected the disease activity in response to ther-

apy.

Hybrid PET/MRI

Rischpler et al. reviewed a comparative summary of the

existing applications for hybrid PET/MRI in the field of car-

diology, and suggested potential cardiac applications that ex-

ploit the unique properties of the newly introduced com-

bined instrumentation (29). Integrated hybrid PET/MRI

analyses allow for the assessment of the quantity of affected

myocardium using LGE, facilitate the assessment of disease

stage and help in treatment decision-making. Moreover, this

technique may be used to assess CS-related inflammatory

cardiomyopathy.

White et al. reported a patient in whom simultaneous

PET/MRI was used to diagnose active CS (30). Interest-

ingly, intrinsic spatial regions of increased 18F-FDG uptake

were found in the subepicardial LGE zone, indicating possi-

ble fibrosis and suggesting inward migration of the inflam-

matory injury. Subsequently, 18F-FDG PET and cardiac MRI

were exploited as a hybrid technique for diagnosing active

CS. Therefore, using cardiac PET/MRI may be clinically

feasible and effective for the detection of inflammatory car-

diac disease.


1602

Echocardiography

Kaderli et al. evaluated the LV function using conven-

tional echocardiography and tissue Doppler imaging in 55

patients with early stage pulmonary sarcoidosis without car-

diac involvement, and compared the findings with those of

22 healthy subjects. The isovolumic acceleration (IVA) is a

measure of LV contractility that is determined by non-

invasive tissue Doppler imaging (31). The IVA was lower in

patients with sarcoidosis than in healthy controls, and the ra-

tio of the myocardial pre-contraction time (PCTm) and con-

traction time (CTm) was higher at the septal annulus (p=

0.026). The IVA values in both the lateral and septal annuli

were also significantly lower in the sarcoidosis group than

in the control group. These observations may reflect sub-

clinical myocardial sarcoid involvement, particularly that of

the interventricular septum. The authors concluded that these

analyses may contribute to clinical decision-making by fa-

cilitating the predictions of cardiac involvement in patients

with pulmonary sarcoidosis.

Focardi et al. evaluated 69 patients with chronic sarcoido-

sis without suspected cardiac involvement and 26 control

subjects using echocardiography, and determined the preva-

lence of LV systolic and diastolic dysfunction in patients

with chronic sarcoidosis without clinical evidence of heart

disease (32). No significant differences were observed in the

atrial size, LV diameter, wall thickness, LVEF or endocar-

dial fractional shortening (FS) between the sarcoid and con-

trol patients. However, the sarcoid patients had lower mid-

wall fractional shortening (mFS). Taken together, these ob-

servations demonstrated the absence of LV systolic dysfunc-

tion, as evaluated by traditional echocardiographic methods,

in patients with chronic sarcoidosis, and an absence of any

relationship between LV diastolic dysfunction and sarcoido-

sis. Only the mFS was lower among sarcoid patients, par-

ticularly in those with a long disease history. Further analy-

ses are required to confirm the significance of this index as

a potential marker of cardiac involvement in patients with

chronic sarcoidosis.

Cytokines

Extensive myocardial sarcoid granulomatous infiltration

results in DCM and heart failure, which can be either sys-

tolic or diastolic in nature. However, studies of autopsy

hearts have demonstrated that granulomatous infiltration is

not distributed diffusely, but instead appears in patches.

Therefore, it is speculated that, in addition to direct injury

from granulomas with myocardial cell loss and interstitial fi-

brosis, other factors, such as proinflammatory cytokines,

may play a role in myocardial dysfunction, particularly in

the acute or subacute phase of CHF in CS patients.

The levels of inflammatory cytokines, including tumor ne-

crosis factor-alpha (TNF-α), interleukin-1β (IL-1β),

interleukin-6 (IL-6) and their soluble receptors, are report-

edly increased in CHF patients (33, 34). Inflammatory cy-

tokines may modulate the myocardial functions through a

variety of mechanisms, including stimulation of hypertrophy

and fibrosis through direct effects on cardiomyocytes and fi-

broblasts, the impairment of myocardial contractile function

through direct effects on intracellular calcium transport and

adrenergic receptor signaling, the induction of apoptosis and

increasing the expression of genes involved in myocardial

remodeling (35).

Type 1 helper T cell (Th1)-related cytokines

The expression of Th1-related cytokines is enhanced by

sarcoidosis. Therefore, we investigated the inflammatory cy-

tokine mRNA expression in the myocardia of 12 CS and 10

idiopathic DCM patients (36). These analyses showed sig-

nificantly enhanced expression of Th1-related cytokines,

such as IL-12p40, IFN-γ and IL-2, in CS patients, and myo-

cardial immunostaining demonstrated that the enhanced ex-

pression of IL-12 was confined to macrophages and giant

granuloma cells. These data suggest that the expression pat-

terns of the Th1-related cytokines IL-12 and IFN-γ may be

differentially diagnostic of cardiac dysfunction in patients

with idiopathic DCM and CS.

TNF-α

TNF-α is known to be a major proinflammatory cytokine

that controls the strength, effectiveness and duration of local

and systemic inflammatory reactions. Numerous studies

have confirmed high TNF-α production in cases with sarcoi-

dosis, and have shown an important role of TNF-α in granu-

loma formation (37-41). Moreover, mononuclear phagocytes,

such as mature macrophages, are responsible for the TNF-αsecretion in patients with sarcoidosis (42). In the study of

the inflammatory cytokines described above, TNF-α mRNA

was expressed in the myocardia of both CS and idiopathic

DCM patients, but it tended to be higher in the sarcoid

myocardia.

Takashige et al. (43) investigated TNFA (TNF-α) and

TNFB (lymphotoxin-alpha) gene polymorphisms in 26 CS

patients of Japanese origin. Significant increases in the

TNFA2 allele were found in the patient group, suggesting

that the TNFA gene is involved in the genetic susceptibility

to CS. Subsequently, HLA-DQB1*0601 was found to be the

allele most significantly associated with CS susceptibility,

and was more significantly increased compared with

TNFA2 (44). Kuroda et al. also analyzed single nucleotide

polymorphisms and showed that the TNF-α-857C/T poly-

morphism may affect the susceptibility to CS (45).

The chimeric monoclonal anti-TNF-α antibody, inflixi-

mab, has been approved for use in patients with rheumatoid

arthritis and Crohn’s disease. Case reports (46-50) have de-

scribed the success of infliximab in patients with sarcoidosis

refractory to conventional therapy. Although the benefits and

indications have not been established in patients with sarcoi-

dosis, these reports suggest that anti-TNF-α therapy may be

effective in patients with certain phenotypes.

Crouser et al. reported five consecutive patients with

CD4+ lymphopenia and assessed the clinical disease mani-


1603

Table　1.　Clinical Characteristics of Patients with CD4+ Lymphopenic Sarcoidosis*

Patient Age Gender Race Disease manifestations Clinical improvement post infliximab

S1 60 M W Lung, LN, hrt, sinus Lung a , hrt b , sinus c

S2 41 M AA Lung, LN, hrt, sinus Lung a , hrt b , sinus c

S3 65 F W Lung, LN, sinus Sinus c

S4 57 M W Lung, LN, skin Skin d

S5 58 M AA Lung, LN, hrt, eyes Lung a , hrt b , eyes e

AA: African American, F: female, hrt: heart; LN: lymph node, M: male, W: white.a A > 10% increase in FVC, total lung capacity, or diffusing capacity was observed.b Initial left ventricular ejection fraction was < 45% with 10 % subsequent increase.c Resolution of chronic sinus symptoms (recurrent infections, congestion) was reported.d Objective improvement in sarcoidosis-mediated skin rash was observed.e Resolution of uveitis was documented.

*Crouser ED, Lozanski G, Fox CC, et al. The CD4+ lymphopenic sarcoidosis phenotype is highly

therapy. Chest 137: 1433, 2010.

Table　2.　Possible Biologic Markers of Sarcoidosis Activity*

Monocyte Origin Neutrophil associated

Angiotensin converting enzyme Collagenase

Lysozyme Elastase

Calcitriol

Kininase Extracellular matrix associated

Thermolysin-like metallopeptidase Procollagen III peptide

Neopterin Hyaluronan

Fibronectin

Lymphocyte associated Vitronectin

Soluble interleukin-2 receptors

Beta 2-microglobulin Others

Adenosin deaminase

*Baughman RP, Costabel U. Markers of sarcoidosis. In: Sarcoidosis.Baughman RP, Ed. Taylor & Francis, New York, 2006: 435-461.

CRP

ESR

Amyloid

festations and peripheral blood T-cell subsets before and af-

ter infliximab treatment (51). Significant increases in the ab-

solute peripheral blood lymphocyte and CD4+ T-cell counts

were observed, and improvements in the clinical disease

manifestations were demonstrated in all infliximab-treated

patients. The authors concluded that the presence of CD4+

lymphopenia identified a distinct sarcoidosis phenotype that

is particularly responsive to anti-TNF-α therapy. Interest-

ingly, in three of five patients, the initial LVEF values were

less than 45% and were subsequently increased by more

than 10% after anti-TNF-α therapy (Table 1).

Other biomarkers

Several biological markers, including ACE, lysozyme and

sIL-2R, have been recognized during assessments of sarcoi-

dosis activity (52) (Table 2). However, the sensitivity and

specificity of those biomarkers are often inadequate for CS,

which requires better biomarkers. Although several new

biomarkers of CS have been reported, none are genuinely

specific to CS. Therefore, integrative diagnoses require im-

aging and histological examinations, although the future di-

agnosis rates may be improved by the use of multiple

biomarkers.

Atrial natriuretic peptide (ANP) and BNP

In a study of 62 patients, Yasutake et al. showed a ten-

dency for there to be increased plasma ANP and BNP levels

in 27 CS patients, and demonstrated the utility of the ROC

analyses of these markers for distinguishing among high-

degree AVB, VT and CHF (53).

Myeloid-related protein complex (MRP 8/14)(S100A8/

A9)

MRP 8/14 is a member of the S100 family of

inflammation-related proteins and is expressed in activated

macrophages and granulocytes. We compared the serum lev-

els of MRP 8/14 in 30 normal subjects, 23 patients with idi-

opathic DCM, 25 sarcoid patients without CS and 10 pa-

tients with CS (54). The serum MRP 8/14 levels were only

significantly elevated in the CS patients, who had enhanced

immunohistochemical expression in macrophages and giant

cells of granulomas from myocardial tissues. Therefore, the

assessment of MRP 8/14 expression may facilitate the diag-

nosis of CS.

High-sensitive cardiac troponin T (hs-cTnT)

Baba et al. evaluated the sensitivity and specificity of hs-

cTnT, BNP, ACE, lysozyme, CMR and Ga-67 scintigraphy

using 18F-FDG PET data as an indicator of sarcoidosis activ-

ity in 12 CS patients (55). The data showed that the sensi-

tivity and specificity of hs-cTnT were 87.5% and 75%, re-

spectively, and the positive and negative predictive values

were 87.5% and 75%, respectively. These investigators con-

cluded that hs-cTnT was an excellent marker for evaluating

the disease activity in CS patients, with superior sensitivity


1604

and negative predictive values and a good responsiveness to

steroid therapy.

High-sensitive cardiac troponin I (hs-cTnI)

Tanada et al. reported the case of a CS patient and deter-

mined the serum hs-cTnI levels over time. Both the serum

hs-cTnI and N-terminal proBNP (NT-proBNP) levels were

increased before the initiation of steroid therapy and de-

creased following treatment (56). While the level of NT-

proBNP was transiently elevated, possibly because of body

fluid retention associated with steroid therapy, the hs-cTnI

levels did not increase; they decreased with the proceeding

therapy. These data indicate that hs-cTnI may indicate myo-

cardial cytopathy in CS patients, and may have potential as

a biomarker of therapeutic responses.

Conclusion

The understanding of CS etiology and disease states has

advanced considerably in recent years, and the diagnostic

and treatment approaches have improved accordingly (57).

However, the temporal changes in cardiac function among

patients with sarcoidosis remain poorly characterized. CHF

usually progresses rapidly within a few weeks or months in

CS patients, and this is sometimes associated with conduc-

tion or rhythmic abnormalities that reflect gross myocardial

granulomatous infiltration. In addition, the influence of in-

flammatory cytokines, such as TNF-α, may be an important

factor in the pathogenesis of progressive congestive cardio-

myopathy in patients with CS. Biomarkers, including high-

sensitive cardiac troponins, may also be predictive of LV

dysfunction.

The observation of both an abnormal uptake on 18F-FDG

PET and abnormal CMR imaging (LGE, and particularly

hyperintensity on T2-weighted images) may therefore make

it clinically possible to detect active inflammation and to

thus predict a deteriorating LV function in CS patients.

However, because histopathological confirmation of myocar-

dial sarcoid lesions cannot be obtained in individual pa-

tients, it remains unclear whether subacute or chronic myo-

cardial sarcoid lesions are accurately reflected by these im-

aging modalities, which may also influence the LV function.

Finally, the establishment of optimal diagnostic and thera-

peutic strategies for CS patients and the preservation of their

LV function remains a challenge, and these topics warrant

thorough prospective multicenter trials.

The authors state that they have no Conflict of Interest (COI).

References

1. Ayyala US, Nair AP, Padilla ML. Cardiac sarcoidosis. Clin Chest

Med 29: 493-508, ix, 2008.

2. Nelson JE, Kirschner PA, Teirstein AS. Sarcoidosis presenting as

heart disease. Sarcoidosis Vasc Diffuse Lung Dis 13: 178-182,

1996.

3. Porter GH. Sarcoid heart disease. N Engl J Med 263: 1350-1357,

1960.

4. Matsui Y, Iwai K, Tachibana T, et al. Clinicopathological study of

fatal myocardial sarcoidosis. Ann NY Acad Sci 278: 455-469,

1976.

5. Roberts WC, McAllister HA Jr, Ferrans VJ. Sarcoidosis of the

heart. A clinicopathologic study of 35 necropsy patients (group 1)

and review of 78 previously described necropsy patients (group

11). Am J Med 63: 86-108, 1977.

6. Chapelon-Abric C, de Zuttere D, Duhaut P, et al. Cardiac sarcoi-

dosis: a retrospective study of 41 cases. Medicine (Baltimore) 83:

315-334, 2004.

7. Smedema JP, Snoep G, van Kroonenburgh MP, et al. Cardiac in-

volvement in patients with pulmonary sarcoidosis assessed at two

university medical centers in the Netherlands. Chest 128: 30-35,

2005.

8. Yazaki Y, Isobe M, Hiroe M, et al. Prognostic determinants of

long-term survival in Japanese patients with cardiac sarcoidosis

treated with prednisone. Am J Cardiol 88: 1006-1010, 2001.

9. Otsuka K, Terasaki F, Eishi Y, et al. Cardiac sarcoidosis underlies

idiopathic dilated cardiomyopathy: importance of mediastinal lym-

phadenopathy in differential diagnosis. Circ J 71: 1937-1941,

2007.

10. Yazaki Y, Isobe M, Hiramitsu S, et al. Comparison of clinical fea-

tures and prognosis of cardiac sarcoidosis and idiopathic dilated

cardiomyopathy. Am J Cardiol 82: 537-540, 1998.

11. Sadek MM, Yung D, Birnie DH, et al. Corticosteroid therapy for

cardiac sarcoidosis: a systematic review. Can J Cardiol 29: 1034-

1041, 2013.

12. Chiu CZ, Nakatani S, Zhang G, et al. Prevention of left ventricu-

lar remodeling by long-term corticosteroid therapy in patients with

cardiac sarcoidosis. Am J Cardiol 95: 143-146, 2005.

13. Guidelines for diagnosis and treatment of myocarditis (JCS 2009):

digest version. Circ J 75: 734-743, 2011.

14. Teramoto K, Shimamoto S, Terasaki F, et al. Temporal changes in

echocardiographic findings in cardiac and non-cardiac sarcoidosis

patients. Intern Med 51: 3001-3007, 2012.

15. Vignaux O, Dhote R, Duboc D, et al. Detection of myocardial in-

volvement in patients with sarcoidosis applying T2-weighted,

contrast-enhanced, and cine magnetic resonance imaging: initial

results of a prospective study. J Comput Assist Tomogr 26: 762-

767, 2002.

16. Schatka I, Bengel FM. Advanced imaging of cardiac sarcoidosis. J

Nucl Med 55: 99-106, 2014.

17. Matoh F, Satoh H, Shiraki K, et al. The usefulness of delayed en-

hancement magnetic resonance imaging for diagnosis and evalu-

ation of cardiac function in patients with cardiac sarcoidosis. J

Cardiol 51: 179-188, 2008.

18. Ichinose A, Otani H, Oikawa M, et al. MRI of cardiac sarcoidosis:

basal and subepicardial localization of myocardial lesions and

their effect on left ventricular function. AJR Am J Roentgenol

191: 862-869, 2008.

19. Watanabe E, Kimura F, Nakajima T, et al. Late gadolinium en-

hancement in cardiac sarcoidosis: characteristic magnetic reso-

nance findings and relationship with left ventricular function. J

Thorac Imaging 28: 60-66, 2013.

20. Patel AR, Klein MR, Chandra S, et al. Myocardial damage in pa-

tients with sarcoidosis and preserved left ventricular systolic func-

tion: an observational study. Eur J Heart Fail 13: 1231-1237,

2011.

21. Yamagishi H, Shirai N, Takagi M, et al. Identification of cardiac

sarcoidosis with 13N-NH3/18F-FDG PET. J Nucl Med 44: 1030-

1036, 2003.

22. Takeda N, Yokoyama I, Hiroi Y, et al. Positron emission tomogra-

phy predicted recovery of complete A-V nodal dysfunction in a

patient with cardiac sarcoidosis. Circulation 105: 1144-1145,


1605

2002.

23. Blankstein R, Osborne M, Naya M, et al. Cardiac positron emis-

sion tomography enhances prognostic assessments of patients with

suspected cardiac sarcoidosis. J Am Coll Cardiol 63: 329-336,

2014.

24. Skali H, Schulman AR, Dorbala S. 18F-FDG PET/CT for the as-

sessment of myocardial sarcoidosis. Curr Cardiol Rep 15: 352,

2013.

25. Morooka M, Moroi M, Uno K, et al. Long fasting is effective in

inhibiting physiological myocardial 18F-FDG uptake and for evalu-

ating active lesions of cardiac sarcoidosis. EJNMMI Res 4: 1,

2014.

26. Mc Ardle BA, Birnie DH, Klein R, et al. Is there an association

between clinical presentation and the location and extent of myo-

cardial involvement of cardiac sarcoidosis as assessed by 18F-

fluorodoexyglucose positron emission tomography? Circ Cardio-

vasc Imaging 6: 617-626, 2013.

27. Mehta D, Lubitz SA, Frankel Z, et al. Cardiac involvement in pa-

tients with sarcoidosis: diagnostic and prognostic value of outpa-

tient testing. Chest 133: 1426-1435, 2008.

28. Matthews R, Bench T, Meng H, et al. Diagnosis and monitoring

of cardiac sarcoidosis with delayed-enhanced MRI and 18F-FDG

PET-CT. J Nucl Cardiol 19: 807-810, 2012.

29. Rischpler C, Nekolla SG, Dregely I, et al. Hybrid PET/MR imag-

ing of the heart: potential, initial experiences, and future pros-

pects. J Nucl Med 54: 402-415, 2013.

30. White JA, Rajchl M, Butler J, et al. Active cardiac sarcoidosis:

first clinical experience of simultaneous positron emission

tomography-magnetic resonance imaging for the diagnosis of car-

diac disease. Circulation 127: e639-e641, 2013.

31. Aydin Kaderli A, Gullulu S, Coskun F, et al. Impaired left ven-

tricular systolic and diastolic functions in patients with early grade

pulmonary sarcoidosis. Eur J Echocardiogr 11: 809-813, 2010.

32. Focardi M, Picchi A, Nikiforakis N, et al. Assessment of cardiac

involvement in sarcoidosis by echocardiography. Rheumatol Int

29: 1051-1055, 2009.

33. Gullestad L, Ueland T, Vinge LE, et al. Inflammatory cytokines in

heart failure: mediators and markers. Cardiology 122: 23-35,

2012.

34. Hori M, Yamaguchi O. Is tumor necrosis factor-alpha friend or foe

for chronic heart failure? Circ Res 113: 492-494, 2013.

35. Mann DL. Inflammatory mediators and the failing heart: past, pre-

sent, and the foreseeable future. Circ Res 91: 988-998, 2002.

36. Terasaki F, Ukimura A, Tsukada B, et al. Enhanced expression of

type 1 helper T-cell cytokines in the myocardium of active cardiac

sarcoidosis. Circ J 72: 1303-1307, 2008.

37. Seitzer U, Swider C, Stuber F, et al. Tumour necrosis factor alpha

promoter gene polymorphism in sarcoidosis. Cytokine 9: 787-790,

1997.

38. Higuchi T, Seki N, Kamizono S, et al. Polymorphism of the 5’-

flanking region of the human tumor necrosis factor (TNF)-alpha

gene in Japanese. Tissue Antigens 51: 605-612, 1998.

39. Zheng L, Teschler H, Guzman J, et al. Alveolar macrophage TNF-

alpha release and BAL cell phenotypes in sarcoidosis. Am J

Respir Crit Care Med 152: 1061-1066, 1995.

40. Sweiss NJ, Zhang W, Franek BS, et al. Linkage of type I inter-

feron activity and TNF-alpha levels in serum with sarcoidosis

manifestations and ancestry. PLoS One 6: e29126, 2011.

41. Baydur A, Alavy B, Nawathe A, et al. Fatigue and plasma cy-

tokine concentrations at rest and during exercise in patients with

sarcoidosis. Clin Respir J 5: 156-164, 2011.

42. Pantelidis P, McGrath DS, Southcott AM, et al. Single-cell analy-

sis: a novel approach to tumour necrosis factor-alpha synthesis

and secretion in sarcoidosis. Eur Respir J 20: 1179-1184, 2002.

43. Takashige N, Naruse TK, Matsumori A, et al. Genetic polymor-

phisms at the tumour necrosis factor loci (TNFA and TNFB) in

cardiac sarcoidosis. Tissue Antigens 54: 191-193, 1999.

44. Naruse TK, Matsuzawa Y, Ota M, et al. HLA-DQB1*0601 is pri-

marily associated with the susceptibility to cardiac sarcoidosis.

Tissue Antigens 56: 52-57, 2000.

45. Kuroda H, Saijo Y, Fujiuchi S, et al. Relationship between cy-

tokine single nucleotide polymorphisms and sarcoidosis among

Japanese subjects. Sarcoidosis Vasc Diffuse Lung Dis 30: 36-42,

2013.

46. Roberts SD, Wilkes DS, Burgett RA, et al. Refractory sarcoidosis

responding to infliximab. Chest 124: 2028-2031, 2003.

47. Baughman RP, Lower EE. Infliximab for refractory sarcoidosis.

Sarcoidosis Vasc Diffuse Lung Dis 18: 70-74, 2001.

48. Pettersen JA, Zochodne DW, Bell RB, et al. Refractory neurosar-

coidosis responding to infliximab. Neurology 59: 1660-1661,

2002.

49. Yee AM, Pochapin MB. Treatment of complicated sarcoidosis

with infliximab anti-tumor necrosis factor-alpha therapy. Ann In-

tern Med 135: 27-31, 2001.

50. Mallbris L, Ljungberg A, Hedblad MA, et al. Progressive cutane-

ous sarcoidosis responding to anti-tumor necrosis factor-alpha

therapy. J Am Acad Dermatol 48: 290-293, 2003.

51. Crouser ED, Lozanski G, Fox CC, et al. The CD4+ lymphopenic

sarcoidosis phenotype is highly responsive to anti-tumor necrosis

factor-α therapy. Chest 137: 1432-1435, 2010.

52. Baughman RP, Costabel U. Markers of sarcoidosis. In: Sarcoido-

sis. Baughman RP, Ed. Taylor& Francis, New York, 2006: 435-

461.

53. Yasutake H, Seino Y, Kashiwagi M, et al. Detection of cardiac

sarcoidosis using cardiac markers and myocardial integrated

backscatter. Int J Cardiol 102: 259-268, 2005.

54. Terasaki F, Fujita M, Shimomura H, et al. Enhanced expression of

myeloid-related protein complex (MRP8/14) in macrophages and

multinucleated giant cells in granulomas of patients with active

cardiac sarcoidosis. Circ J 71: 1545-1550, 2007.

55. Baba Y, Kubo T, Kitaoka H, et al. Usefulness of high-sensitive

cardiac troponin T for evaluating the activity of cardiac sarcoido-

sis. Int Heart J 53: 287-292, 2012.

56. Tanada Y, Sato Y, Sawa T, et al. Serial measurement of high-

sensitivity cardiac troponin I and N-terminal proB-type natriuretic

peptide in a patient presenting with cardiac sarcoidosis. Intern

Med 51: 3379-3381, 2012.

57. Nunes H, Freynet O, Naggara N, et al. Cardiac sarcoidosis. Semin

Respir Crit Care Med 31: 428-441, 2010.

Ⓒ 2014 The Japanese Society of Internal Medicine

http://www.naika.or.jp/imonline/index.html

deterioration of cardiac function during the progression

Documents