Sarcoidosis is a multisystemic granulomatosis. The prevalence of this disease varies between 10 and 20/100,000/ per year47,101 in whites and 35/100000/ per year in African Americans101. Lymph node, skin, lung infiltration, cutaneous infiltration, and ophthalmologic manifestations are the most frequent signs. Cardiac sarcoidosis (CS) is noted in 3%-17% of cases in clinical series18,79,114, and in 20%-30% of cases in pathology series52,94,99,114. Recent studies indicate an increasing number of cases of CS and a possible racial factor. For Japanese females53,105, the occurrence of CS is associated with the presence of HLA-DQB1*06085 and the allele TNFA2122.
Responsible for 50% of the deaths of patients with sarcoidosis52,94, myocardial infiltration is all the more severe and diffuse where there are clinical and/or electrical signs99,114. Infiltration is most frequently identified in the intraventricular septum and/or in the left ventricle34,71,99,114. Involvement of the right ventricle, atrial wall, and papillary muscles is rarer, as well as any pericardial involvement26,34,108. Exceptionally, an infiltration of media and adventitia of the small coronary vessels78 and involvement of valvular cusp18,28,34,99 can be found.
Because of 1) the number of cases of CS, 2) the variety of ethnic groups in the study, 3) the variety of clinical expressions, and 4) the prolonged follow-up, the current study is the first, to our knowledge, to allow a fundamental appreciation of the clinical and laboratory parameters of CS, its relationship with other localizations, its prognosis, and therapeutic approaches.
PATIENTS AND METHODS
This cooperative study undertaken by 3 departments of internal medicine concerns 41 cases of CS, observed over 15 years. All patients corresponded to the following diagnostic criteria55:
- a suggestive or compatible clinical and radiologic picture;
- the absence of exposure to an organic or inorganic material capable of inducing granulomatosis, particularly professional exposure to dust particles;
- the absence of elements for tuberculosis, fungal mycosis, and parasitic disease; and
- the presence of noncaseating granulomas composed of multinucleated giant cells surrounded by lymphocytes.
The diagnosis of cardiac involvement was determined according to 3 different circumstances and depending on the presence of clinical and laboratory signs:
- Symptomatic cardiomyopathy due to sarcoidosis was diagnosed when patients presented cardiac dyspnea and at least 1 laboratory abnormality;
- CS was diagnosed when patients showed simultaneous functional cardiac signs (dyspnea excluded), such as thoracic pain, syncope, and at least 1 abnormal laboratory exam;
- Finally, in a few patients, the diagnosis of CS was based on the combination of compatible echographic and scintigraphic abnormalities, together with a favorable outcome with corticosteroids. Any other pathology has been excluded.
The literature led us to identify the last group based on laboratory signs. Autopsy series14,48,67,69 have shown correlations between echographic and/or scintigraphic findings and clinical signs. These explorations permit the diagnosis of CS that can be considered relatively mild owing to the absence of clinical and electrical signs114.
The diagnosis of CS was retained when there was at least a clinical electrical sign, echocardiographic abnormalities, and/or either defect with the radionuclide studies (sestamibi or thallium-201 imaging at rest, during exercise, or after dipyridamole test). No patient presented hypertensive, valvular, or ischemic heart disease, which might be misleading for the interpretation of cardiac results. None was a smoker or had diabetes or dyslipidemia.
Extracardiac manifestations and their respective courses were compiled for all patients, together with the modalities of apparent cardiac signs with regard to the diagnosis of sarcoidosis (before, during, or after extracardiac signs). CS was estimated according to the presence of functional or physical signs, the New York Heart Association (NYHA) classification, the results of the electrocardiogram (ECG), 24-hour Holter monitoring, echocardiography, radionuclide studies, and, in a few cases, cardiac catheterization. Functional respiratory investigations and diffusing capacity (DLCO) measurement were performed to estimate the pulmonary infiltration.
The course of the CS was evaluated at 6 months in 41 patients and at more than 1 year in 39 patients. Based on the clinical study (NYHA classification) and laboratory results (ECG, 24-hour monitoring, echocardiography, and radionuclide imaging), we defined each patient's case as 1 of the following categories: cure without residual cardiac signs, improvement with residual laboratory signs, stabilization, or worsened condition.
Data analysis was performed with SAS (Statistical Analysis System, SAS Institute Inc., Cary, NC). A chi-square test or a Fisher exact test was used to compare proportions, and a 95% confidence interval was computed for each odds ratio when applicable. A Wilcoxon rank sum test was computed for ordinal variables and for continuous variables not normally distributed. An analysis of variance was performed for normally distributed continuous variables.
The current study concerns 18 female patients and 23 males (sex ratio F/M, 0.78). Patients were white in 73% of cases and black African or Caribbean natives in 27% of cases. The average age at onset of extracardiac sarcoidosis and at diagnosis was 36 years (range, 6-63 yr); the average age at CS diagnosis was 38 years (range, 18-66 yr). The average age, although not significant, varied according to sex: for females it was 40.2 years (range, 24-62 yr) and for males, 36.4 years (range, 18-66 yr) (not significant [NS]). This parameter also varied according to ethnic groups, the oldest being white patients, respectively 39.6 years (range, 18-66 yr) versus 33.6 years (range, 23-40 yr) for black patients (NS).
Extracardiac signs are shown in Table 1. All patients presented with an extracardiac symptom at least once during the course of the disease. Constitutional symptoms (fatigue, poor general condition, weight loss >5 kg, and/or temperature >38°C) were present in 23 patients (56%). A thoracic localization, symptomatic in 1 patient, was noted in 90% of cases. All different radiologic stages were observed. Thirteen of 39 (33%) patients studied had an abnormal pulmonary function test: 2 obstructive patterns and 11 restrictive syndromes. Decreased diffusing capacity (DLco) was noted in 46% of cases. Bronchoalveolar lavage was performed in 30 patients and revealed a high lymphocytosis (T4) in 83% of cases. Tuberculin skin reaction was negative in 95% of cases. There was a high frequency of neurologic sarcoidosis (27% including 1 or more involvements: cerebral nervous system involvement in 7 cases, cranial nerve palsies in 2 cases, muscular involvement in 2 cases, and peripheral polyneuropathy in 2 cases).
The erythrocyte sedimentation rate was greater than 50 mm/h in 27% of cases. An abnormal blood cell count was noted in 5 cases (12%): eosinophilia, 3; elevated leukocytes, 1; thrombocytopenia, 1. Elevated serum alkaline phosphatase levels were noted in 11 cases (27%). Hypercalcemia and hypercalciuria were present in 7% and 12% of cases, respectively. The serum angiotensin converting enzyme, measured in 38 patients, was elevated in 79% of cases.
Morphologic Examinations and Radionuclide Imaging
Gallium-67 scans, performed in 20 patients, showed multiple disseminated lesions in 85% of cases and an abnormal cardiac fixation in 2 cases. A thoracoabdominal scan confirmed thoracic localization and revealed abdominal lymph nodes in 1 case. Ten patients were evaluated by cerebral computerized tomographic (CT) scan and/or magnetic resonance imaging (MRI), which demonstrated abnormal enhancements in 5 of 7 patients with central involvement.
All patients had histologic demonstration of sarcoid tissue before treatment. More than 1 biopsy was required for 32 patients (Table 2): 2 sites in 20 cases, 3 sites in 7 cases, and 5 sites in 1 case. Noncaseating granulomas were present in 75% of 74 biopsies and 3 Kveim tests performed.
CS appeared before systemic sarcoidosis in 4 patients; the average time for diagnosis was 29 months (range, 6-48 mo). Cardiac signs were associated with constitutional symptoms in 11 cases (27%). Finally, in 26 cases (63%), cardiac signs appeared an average of 4 months after the diagnosis of sarcoidosis (extreme, 6-18 mo). In 6 of these 26 cases, cardiac signs appeared after the corticosteroid dose was reduced.
Presenting manifestations of CS were a functional sign in 18 patients (44%), an abnormal auscultation in 8 cases (19%), and electrical signs in 9 cases (22%). Finally, in 6 cases (15%) of severe multisystemic sarcoidosis and/or neurosarcoidosis, the presence of CS was searched for systematically. Echocardiographic abnormalities (4 cases) and defect of thallium-201 scan (2 cases) revealed the presence of CS in those cases.
Clinical signs were noted in 18 cases (Table 3). Seventeen patients (41%) had progressive dyspnea. At diagnosis, only 1 patient showed a symptomatic pulmonary infiltration (radiologic stage IV). Stage II of the NYHA classification was the most frequent (32%). Signs of left ventricular insufficiency were observed in 5 cases, associated with right ventricular insufficiency in 4 cases. Three patients (7%) had brief syncopal manifestations. An abnormal auscultation was noted in 14 patients (34%): irregular rhythm in 7 cases (17%); cardiac murmur in 5 cases (mitral 4, tricuspid 1, and aortic 1); and gallop in 5 cases.
The ECG was abnormal in 28 patients (68%) (Table 4), and 5 patients (19%) had several abnormalities. In these 28 patients, conduction disturbances were present in 43% and arrhythmia in 46%. Abnormal T waves, signs of left ventricular hypertrophy, and abnormal R waves were noted in 7%, 11%, and 7% of the 28 cases, respectively.
Late ventricular potential was searched for in 4 patients and proved positive in 2 cases (50%). Holter ECG monitoring, performed in 14 patients, was abnormal in 70% of cases. Compared with basal ECG, it confirmed ECG data in 5 cases and revealed a more severe conduction disturbance in 2 cases and a ventricular arrhythmia in 1 case.
An electrophysiologic study was performed in 13 patients. Results confirmed ECG abnormalities in 9 patients (69%) and identified the site of the block as nodal (4 cases), suprahisian (2 cases), and infrahisian (2 cases). Finally, sinus dysfunction was noted in 2 cases. A ventricular stimulation program, used in 3 patients, provoked ventricular tachycardia in 2 cases. Two patients had associated abnormalities: a conduction disturbance at 2 levels (1 case) and arrhythmia with a conduction abnormality (1 case).
The cardiothoracic ratio was elevated in 7 patients (17%), with an average at 0.63, independently of a pericardial involvement. Doppler echocardiography, performed in 40 patients, was normal in 9 cases (23%). Various abnormalities, sometimes multiple, were noted in 15 patients (Figure 1; Table 5). Among these, the following were dominant: an aspect of dilated cardiomyopathy (32%), abnormal left ventricular relaxation (29%), and diffuse or localized dyskinesia or hypokinesia (26%). Valvular murmurs were correlated with valvular regurgitation in 2 cases, with abnormal cups.
A sestamibi or thallium-201 scan was performed in 40 patients either at rest (17 cases), during exercise (5 cases), or after dipyridamole test (18 cases). Radionuclide imaging was normal in 10 cases (25%) (at rest: 6 cases, dipyridamole test: 4 cases). A defect was observed in 30 patients: patchy thallium perfusion defect in 19 cases (75%), localized defect in 6 cases, and diffuse perfusion defect in 5 cases. In all cases but 1, defects were regressive with dipyridamole or during the redistribution and did not deteriorate with exercise.
Cardiac catheterization (4 cases) and isotopic cardiac output (2 cases) confirmed the wall motion abnormalities already noted during echocardiography. In 1 case, a coronary angiogram proved normal.
Regarding the above-mentioned data, comparative studies did not show significant clinical or laboratory differences according to sex and ethnic group. Nevertheless, patients over 40 years had significantly more atrioventricular block and ventricular arrhythmia (p < 0.05), whether or not they presented with symptomatic cardiomyopathy.
The diagnosis of CS was made easily in 17 patients due to the coexistence of physical signs (dyspnea) and at least 1 laboratory abnormality (ECG, echocardiography, radionuclide imaging) (Figure 2). For 7 patients who were symptomatic but without dyspnea (Figure 3), diagnosis was made based on clinical findings-in 1 case, the presence of associated electrical signs. In 6 others cases, clinical signs were associated with 2 or 3 laboratory abnormalities. In 17 asymptomatic patients (Figure 4), the diagnosis of CS was made based on the presence of different criteria: ECG signs and echocardiographic abnormality in 3 cases; ECG signs and a thallium-201 defect in 3 cases; ECG, echographic, and radionuclide imaging abnormalities in 4 cases; and echocardiographic and thallium-201 defect abnormalities in 3 cases. Finally, in 4 patients, diagnosis was made based on the presence of either an echographic abnormality (2 cases) or a thallium-201 defect (2 cases) that arose in the context of systemic sarcoidosis and responded to therapy (Table 6).
These laboratory differences led us to investigate various examinations and their combined results (Table 7). A normal ECG could be combined with echocardiographic abnormalities (13 cases), such as localized hypokinesia (4 cases), left ventricular diastolic dysfunction (3 cases), pericardial abnormality (5 cases), cardiomyopathy (1 case), and a septal thickness (1 case). Conversely, ECG abnormalities, such as arrhythmia (3 cases) and/or conduction disturbance (3 cases) or left ventricular hypertrophy (1 case), were not necessarily associated with echocardiographic abnormalities (7 cases).
A normal ECG did not eliminate the possibility of thallium perfusion defect. Conversely, electrical abnormalities might not be correlated with radionuclide imaging abnormalities (see Table 7), in particular during arrhythmia (4 cases) or with conduction disturbance (2 cases).
Echocardiography was normal in 8 cases (see Table 7), despite the presence of indisputable thallium defects: patchy (4 cases) or localized (4 cases). Conversely, a normal thallium-201 scan was observed while echocardiography was pathologic in 9 patients (see Table 7): restrictive cardiomyopathy (3 cases), diastolic dysfunction (3 cases), localized hypokinesia and pericardial effusion (2 cases), and left ventricular aneurysm (1 case). Finally, we found no significant association between the results of ECG, echocardiogram, and radionuclide imaging considered 2 by 2, respectively (see Table 7).
Due to the retrospective character of our study, there was no precise therapeutic protocol. Two patients did not receive any corticosteroids. One had a small asymptomatic pericardial effusion and a complete heart block treated with a permanent cardiac pacemaker.
Thirty-nine patients received prednisone. The initial daily dosage was determined by the severity of the CS (36 cases) and/or noncardiac localization (3 cases): 0.25 mg/kg (2 cases), 0.5 mg/kg (12 cases), 0.5 then 1 mg/kg (5 cases), 1 mg/kg (19 cases), and 1.5 mg/kg (1 case), for 6-8 weeks. In 8 cases, because of severe cardiac (6 cases) and noncardiac (2 cases) sarcoidosis, intravenous methylprednisolone was administered (10-15 mg/kg for 3 consecutive days). An improvement with regression of signs was quickly observed in 31 patients, allowing for a progressive decrease of the daily dosage which, over several months, led to a maintenance dosage of below 10 mg/day. The average duration of steroid therapy was 43 months (range, 6-168 mo).
One or several immunosuppressive therapies were added for 11 patients: cyclophosphamide (8 times with a monthly bolus of 500-700 mg/m2; methotrexate (6 times with weekly intramuscular injections of 20-30 mg); or cyclosporin (3 times with doses between 3 and 5 mg/kg per day giving a radioimmunologic dosage between 80 and 180 ng/mL) (Table 8). The selection of these agents is described below. Corticosteroids were associated with a combination of 3 immunosuppressive agents in 1 patient, of 2 agents in 3 patients, and of 1 agent in 7 patients.
Cardiac treatment was prescribed in 10 cases: beta blockers (2 cases), angiotensin-converting enzyme inhibitors (1 case), antiarrhythmic therapy (3 cases), calcium channel blockers (1 case), diuretics (5 cases), digitalics (3 cases), and a permanent pacemaker (1 case). Finally, a tardily treated patient who received immunosuppressive therapy (prednisone plus methotrexate) that proved ineffective underwent a heart transplantation because of a worsened cardiomyopathy (in stage IV of NYHA) and resistance to cardiac treatment.
The course at 6 months was marked by normalization or improvement of clinical and/or laboratory parameters in 90% of 41 patients; stabilization was obtained in 2 cases (5%); and 2 patients worsened.
Long-term outcome, beyond 6 months, was judged in 39 patients (2 patients were lost to follow-up after 4 and 5 months, respectively). The average duration of follow-up was 86 months (range, 10-312 mo) for systemic sarcoidosis and 58 months (range, 7-312 mo) for CS.
Extracardiac localization evolved favorably except for neurosarcoidosis, which was responsible for a residual deficit in 3 cases. The follow-up was marked by 6 deaths: 1 immediately after heart transplantation (unknown cause) and 5 deaths not connected with sarcoidosis, an average of 33 months after the diagnosis of CS (range, 24-56 mo). The causes of the 5 unrelated deaths were digestive in 2 cases, suicide in 2 cases, and lymphoma in 1 case. No patient died from sudden death.
Improvement was noted in 87% of the 39 cases, with complete recovery (clinical and/or laboratory) in 21 patients (54%) and improvement with clinical or laboratory residual deficit in 13 patients (33%). Clinical or echocardiographic stabilization was noted in 3 patients (8%). Finally, 2 cases (5%) worsened with an unfavorable course of cardiomyopathy: in the first case, the immunosuppressive therapy (corticosteroids and methotrexate) was proposed too late (4 years after the onset of cardiac symptoms); in the second case, the patient benefited from a pacemaker but CS kept progressing.
Response to immunosuppressive therapy was complete recovery in 4 of 11 patients, improvement in 6 patients, and unfavorable response in 1 case. Whether in the initial phase or during recurrence, the response was variable: methotrexate was effective 3 times out of 6; cyclophosphamide, 6 times out of 8; cyclosporin, 2 times out of 3 (see Table 8). In 1 case, the combination of prednisone, cyclophosphamide, and methotrexate failed, but the patient improved later with cyclosporin; another patient improved with methotrexate after the failure of prednisone and cyclophosphamide.
During the follow-up, 9 patients relapsed. All had received corticosteroids (average duration of prednisone, 38 mo; range, 12-126 mo). Upon relapse, the prednisone dose was equal to or lower than 10 mg/day in 5 patients and 25 mg/day in 1 patient; it had been stopped 3-6 months previously in 3 cases. For those patients who relapsed, the mean age at diagnosis of CS was 41.2 years. Patients were white in 78% of cases, with a male:female sex ratio of 0.8. Moderate recurrences was treated by an increase of prednisone in 3 cases; the addition of cyclophosphamide in 3 cases; and of methotrexate in 3 others. All evolved favorably: 6 patients were cured and 3 showed clinical and laboratory improvements. An extracardiac relapse was observed in 3 patients either during the tapering of prednisone (2 cases) or 2 months after its withdrawal (1 case).
Treatment was stopped in 13 patients who were apparently cured, after a mean duration of corticosteroids of 34 months (range, 9-109 mo). During a 36-month follow-up (range, 4-92 mo), 3 relapses were observed, affecting the heart in 2 cases. No clinical or laboratory feature was significantly associated with an increased risk of relapse.
We researched predictive factors for good or poor outcome. Although the course was globally satisfactory, it differed according to the clinical expression and/or laboratory parameters of CS (Figure 5). A total cure of atrioventricular block, confirmed with successive ECGs and Holter monitor recording, was noted in 75% of cases. A normalization of echocardiography and/or radionuclide imaging was obtained in 66% of cases. A cure of arrhythmia was observed in only 55% of cases; residual abnormalities essentially concerned auricular fibrillation. Among the 17 patients with symptomatic cardiomyopathy, 82% evolved favorably, with a clinical and laboratory cure in 47% of cases. Concerning the course of the NYHA classification, the disappearance of all cardiac signs was observed in 53% of cases and an improvement of the functional class was seen in 2 cases (12%) (Figure 6) (p < 0.01). Stabilization or degradation of cardiac insufficiency occurred in 3 and 2 cases, respectively. In asymptomatic cardiomyopathy, residual deficits were observed: electrical signs (3 cases), echocardiographic signs (7 cases), and/or radionuclide abnormalities (7 cases). Patients who had only echocardiographic and/or radionuclide abnormalities became normal in 66% of cases (4 cases).
Although total regressions were observed mostly in patients with atrioventricular block and treated with corticosteroids, the search for factors predictive of a cure showed that an atrioventricular block was a bad predictive factor (p < 0.05). Age older than 40 years, being female or white, atrial arrhythmia, or localized radionuclide defects were not significant pejorative factors.
Treatment-related side effects observed during the follow-up were obesity (4 cases), hepatic methotrexate toxicity (2 cases), bilateral osteonecrosis of the femoral heads (1 case), cataract (1 case), diabetes (1 case), and infectious disease (1 case). Cerebral infarctions due to severe dilated cardiomyopathy, without primary hemostasis disorder, were observed in 2 cases. Lymphoma occurred in 2 patients treated only with prednisone.
According to clinical (3%-17%)18,80,114 or autopsy series (12%-30%)52,114, systematic electrical search leads to suspicion of CS in 9%-51% of cases114,116. All these previous studies did not take into account echocardiography and radionuclide imaging. These 2 examinations fully justified, in the current series, the inclusion of patients having systematic active sarcoidosis, developing in young subjects with no associated pathology or risk factors that could induce heart disease. We were thus able to study 41 cases of CS.
The current series is, to our knowledge, the first large study providing extensive data on clinical and major laboratory investigations. For the most part, patients in our series presented with a relatively mild form of the disease without clinical or electrical signs. Other reports have focused on either clinical-anatomical data34,53,94,99 or investigations such as ECG and 24-hour Holter monitoring119, echocardiography14,69, or radionuclide imaging48,67.
In the current series, as in that of Fleming36, most patients were white (61%). This is in contrast to the main autopsy studies53,94,99 that observed CS preferentially in African American and Japanese patients. Usually, incidence is weaker in white patients (14%) compared with that noted in African American patients (21%) and especially in Japanese patients (68%)53,78. This difference is even higher when age and sex are taken into account, and can affect 78% of the cases of sarcoidosis in Japanese women more than 50 years old78. Histologic expression also varies according to ethnic groups, with extensive myocardial infiltration seen in Japanese patients, while African American and white patients have a focused disseminated granulomatosis78.
The age of patients having "classic" sarcoidosis (without rare localizations) usually varies between 20 and 40 years47, with a second peak for women over 50 years in all ethnic groups. In the current series, diagnosis of sarcoidosis occurred with a mean age of 36 years. CS was observed later, at a mean age of 38 years, (in the older group, 22% were white postmenopausal women). In large series, notably autopsy studies which mainly concerned Japanese or African American patients, the average age of CS was, in most cases, above 45 years53,94,99,114,144. In the current series, women were older than men, although not significantly; as for whites, both males and females were older than blacks. Unlike the literature35,53,94,99,114,144 we did not notice any predominance of female patients.
CS can be evoked in various circumstances taking into account the onset with regard to systemic sarcoidosis, and, also, its clinical or laboratory presentation. In the current series, 63% of the CS were observed after sarcoidosis had been diagnosed. The average time between systemic signs and cardiac signs was brief (4 mo). In 23% of cases, cardiac signs appeared during the reduction of corticosteroids. During early relapses, the diagnosis of CS does not raise any difficulties. When cardiac signs occur 12-180 months after the diagnosis of systemic sarcoidosis (18 cases in the current series), while sarcoidosis is considered cured, an accurate evaluation is necessary to eliminate an infectious, inflammatory, metabolic, or tumoral myocarditis. This diagnostic attitude was equivalent to the one we adopted when CS was the presenting manifestation of sarcoidosis (10% of our series), explaining an average diagnostic delay of 29 months. The later occurrence of systemic signs of sarcoidosis and the retrospective analysis lead to a diagnosis of CS22.
When CS arises simultaneously with other systemic signs (11 cases), it can be either revealing (1 case) or noted in the evaluation of granulomatosis. The simultaneity of signs allows a quick diagnosis. In our series the CS was always associated, sooner or later, with noncardiac localization, unlike neurosarcoidosis, which can remain totally isolated20.
In the literature, CS, which is sometimes revealing, mostly appeared simultaneously with the other classic localizations of the disease42,71,117,137. Our experience leads us to insist on examining patients regularly, notably during the reduction of corticosteroids. During this time, clinical and electrical examinations must be performed regularly, every 3 months.
Constitutional symptoms, mostly moderate, were present in 56% of cases. Although they have no diagnostic value, they must be searched for, because their presence leads to questioning of the "idiopathic" character of a heart disease. Tuberculin skin reaction was negative in 95% of cases.
Unlike Yazaki et al144, we observed mediastinal lung involvement in 90% of our cases. In 40 patients, this localization was not symptomatic, allowing a clean interpretation of the dyspnea. In the majority of cases, observed restrictive syndromes were moderate, as was the decrease in diffusing capacity (DLco). The CT scan always allowed an appreciation of the degree of mediastinal lung involvement. Our results concerning bronchoalveolar lavage are in agreement with published data, which revealed a classical lymphocytosis in 83% of cases136.
The multisystemic character of this disease is illustrated by the variety of localizations observed in our patients, with a higher frequency in joints, parotids, nasal mucosa, and bones. We did not observe cutaneous lesions, notably facial annular lesions, as recently indicated by Okamoto et al88 in a Japanese series. A semiologically interesting point, already mentioned by Lower et al72, was the high frequency of neurosarcoidosis, noted in 27% of our cases. CS arose either simultaneously and hardly symptomatic, or secondary to the decrease of corticosteroids. These results lead us to propose for all patients with neurosarcoidosis, cardiac investigations including ECG, echocardiography, and radionuclide scan, which must be renewed in the follow-up period, particularly during the reduction of corticosteroids, and in the case of any clinical or electrical cardiac sign.
The results of laboratory evaluations were consistent with sarcoidosis, with inconstant abnormalities of erythrocyte sedimentation, blood cell counts, serum and urine calcium levels, and hepatic parameters. Elevated serum angiotensin converting enzyme was noted in more than 75% of cases, attesting to the activity of the disease118. The gallium-67 scan, performed in 20 patients, demonstrated the activity in 85% of cases, including an abnormal cardiac uptake in 2.
All patients had histologic confirmation of sarcoidosis. In 32 patients (78%), at least 2 biopsies of different organs were performed. This series confirms our previous experience19; namely the necessity of performing biopsies on several sites and the usefulness of investigating clinically affected organs (pleural or pericardial tissues, mediastinal lymph node, lung, bone, lachrymal gland).
In the current series, myocardial biopsies were done in 1 case only, and proved negative. Although necessary for the diagnosis, in association with clinical and laboratory results4,7,10,50,60,71,82,87,96,104,128,130 this invasive investigation does not seem justifiable to us, due to its low positivity. In the 1999 study of Uemura et al128, only 20%-30% of endomyocardial biopsies were positive, despite analysis of 4 different myocardial sites in 19% of cases. This is due to the mosaic disposal of the lesions, and their preferential left ventricular or septal localization105,130 distinct from the usual sites of myocardial biopsy96. Therefore several authors suggest using the results of echocardiography, notably presence of a change of the telediastolic right ventricular pressure96; radionuclide imaging; and MRI29,98 to obtain multiple samples for a comprehensive tissue analysis. To this lack of sensitivity must be added a lack of specificity, which depends strictly on reserved histologic criteria96, notably regarding the diagnosis of myocarditis with multinucleated giant cell39.
Mode of Onset
The circumstances of discovery are variable among series. Even today, diagnosis is frequently made at autopsy52,94,99,105,114,134. The causes of death are either cardiac, notably sudden death in 17% of cases6,14,35,97,99,131,134, or extracardiac52,94,99,105, mainly through lung involvement, with fortuitous discovery of CS.
In the current series revealing cardiac signs were functional signs in 44% of cases, abnormal auscultation in 19%, and abnormal investigation in 37%. In 85% of cases, interrogation, clinical examination, and ECG led to detection of the cardiac localization. In 15% of cases, the severity of sarcoidosis, notably the presence of neurologic manifestations, led us to look for possible heart involvement through echocardiography and radionuclide imaging.
In the literature, the revealing cardiac signs were largely dominated by sudden death, noted in 23%-66% of cases1,14,34,43,60,95,99,114, while cardiac insufficiency was revealing in only 5% of cases99,114. The severity of signs described in the literature most probably comes from more severe CS. Very exceptionally, an acute corticosensitive mitral insufficiency has been described81 as well as pseudomyocardial infarction61, acute pericarditis62, pericardial effusion51, tamponade62,100,148, cardiogenic shock82,117, or mimicking right ventricular dysplasia92,113. It is noteworthy that in most cases all these signs arose in young patients generally not subject to ischemic or hypertensive heart diseases.
Revealing or not, sudden death is among the main manifestations, which shows the gravity of CS1,14,34,43,64,95,99,114,131,141. In the current series, no sudden death was observed. The mechanisms are a complete atrioventricular block64,129 by 1) destruction of the atrioventricular node99; 2) granulomatous angiitis of the sinus artery13; 3) infiltration of the sinus node1. Sudden death can be due to a severe ventricular arrhythmia provoked by granulomatous infiltration of the ventricle60,114 or an asystolic arrest75. The risk of sudden death is extremely high where there is severe and massive myocardial infiltration114.
Functional or physical signs of ventricular dysfunction are described, in more than a third of cases7,44,60,90,95,99,114. As in the current series, dyspnea was progressive and not severe. More than 75% of our patients were in stage II of the NYHA. The interpretation of the dyspnea needs to take into account the underlying respiratory involvement. In most cases, the signs of cardiac insufficiency are the illustration of a restrictive cardiomyopathy42,99. This cardiomyopathy is the expression of a granulomatous ventricular infiltration that is responsive to corticosteroids, or of a fibrosis unresponsive to immunosuppressive agents and cardiac therapeutics9,44,50,78,95,99,130. Arrhythmia9, conduction disturbances9,15,42, and strokes can reveal CS140. We observed these complications in 12%, 35%, and 2% of cases, respectively.
The study by Yazaki et al146 comparing CS and idiopathic dilated cardiomyopathy points out certain distinctions. In that study, CS is essentially observed in females and presents with a high incidence of right bundle branch block, while atrial arrhythmia is significantly less frequent than in idiopathic dilated cardiomyopathy. In our study, the female:male sex ratio of the 17 symptomatic cardiomyopathy cases was 0.8. We noted 1 case of atrial flutter and only 1 case with a right bundle branch block. These different results could be due to the age of our patients (younger than 40 years in the current study versus 53 years in the study by Yazaki et al) and to ethnic differences (no Japanese patients in our series).
Hypertrophic cardiomyopathy79,129,145 and acute myocarditis, sometimes fatal117, were described anecdotally in the literature.
Palpitations, lipothymia, and syncope bear witness to arrhythmia or abnormal conduction. Syncope, observed in the current series, led to the discovery of atrioventricular block in 2 cases and a ventricular arrhythmia in 1 case. Angina chest pains can occur, sometimes associated with electrical abnormalities9,111. These phenomena could be connected with a transmural invasion of the left ventricle135. In the series reported by Wait et al135, thoracic pains (28% of cases) were observed in patients with myocardial defects on radionuclide imaging, corresponding to microspasms, without abnormal coronary arteries4. In our series, no patient presented with thoracic pains. The only coronary angiography performed ruled out coronary artery disease.
During the clinical examination, valvular dysfunctions were constantly related to a functional incompetence of the mitral or tricuspid valve. Aortic insufficiency observed in 1 patient without valvular echocardiographic abnormality remained unexplained, as well as the regression under treatment. Organic valvulopathies are exceptional11,18,28,36,99,110, unlike mitral incompetence due to papillary muscle involvement14,62,135.
Pericardial involvement is observed in fewer than 10% of cases9,44,69,78,95,108, and remains essentially asymptomatic. This percentage amounts to 20% during systematic echocardiography62. Pericarditis is mostly associated with cardiomyopathy51. In the current study, effusions (5 cases) as well as bright pericardial echoes (1 case) were asymptomatic. Only 1 patient had underlying myocardial involvement. In the literature various clinical forms have been described: tamponade100,132, acute pericarditis62, relapsing pericarditis95,99,108, pericardial effusion3,51, and at times massive5,25,51,62 and constrictive chronic pericarditis95. When a thoracotomy is performed, the liquid is mostly serosanguineous62,132, exceptionally hemorrhagic44, and pericardial biopsies51,108 are required.
Finally, patients can have clinical and/or electrical manifestations in relation with a ventricular aneurysm8,45,54,74,76,110. In 1 of our patients, the left ventricular aneurysm appeared before treatment and was revealed by ventricular tachycardia. In the literature, left or right ventricular aneurysms of variable size arose more often before steroid therapy8,45,68,74,76; they were sometimes short term8 and were mostly associated with cardiomyopathy. They can be responsible for resistant ventricular arrhythmia45,74, cardiac insufficiency35, or sudden death35.
Although ECG is a key exam, it remains poorly sensitive, revealing electrical abnormalities in only half of the cases114,116. In 7 patients of our series, a systematic ECG performed in the initial phase prompted the decision to pursue the cardiac investigations. The abnormalities observed in the current series are those usually described in the literature, mainly arrhythmias9,34,42,73,78,86,88,99,114,116 and conduction abnormalities34,42,44,86,95,99,114. However, we noted certain differences that deserve to be emphasized and may be connected with the predominant white ethnic background of patients in the current study. Atrial arrhythmias were more frequent than ventricular ones (33% vs. 11% of cases with abnormal ECG, respectively), which probably explained the absence of sudden death. We did not note any Q wave. Finally, complete atrioventricular blocks were infrequent.
At rest, 4 of our patients had sinus tachycardia. Whether present at rest12,41,66 or during exercise41, tachycardia is frequently associated with an isotopic cardiac dysfunction41. These 4 patients had either echocardiographic (4 of 4 patients) and/or radionuclide (1 of 4 patients) abnormalities.
Atrial fibrillation and flutter, observed in our series (3 cases), are exceptional99. None of the patients had echographic atrial dilation or an abnormal ventricular left function. The pathogenic hypothesis was an atrial granulomatous infiltration. This was strengthened by the recovery under immunosuppressive therapy99,114. Ventricular arrhythmias were noted in only 3 of our patients, while they represent more than 50% of electrical abnormalities in other series9,31,42,73,78,87,99,114. Ventricular ectopic beats have been reported in more than a third of cases34,114, sometimes leading to sustained tachycardia9,44,60,95,99,121 or even ventricular fibrillation73,78,87,122 at times responsible for sudden death.
Heart blocks represent one of the most privileged expressions of CS and were observed in 47%-91% of cases34,42,44,95,99,114,119 whether they concerned intraventricular blocks, in particular of the right branch34,52,86,99,119,127,147, or atrioventricular blocks9,34,38,99 and even sinus blocks1,21,44,105. The complete heart blocks represent one of the mechanisms of sudden death. For Japanese patients, notably women over 50 years129,147, complete heart blocks are frequent and lead to the discovery of CS in 11% of cases. Our series illustrates the various conduction abnormalities, but unlike the series by Sekiguchi et al105, atrioventricular blocks were associated with relapses and a less favorable course of CS, although taken individually they could be cured in 75% of cases.
The diffuse character of the infiltration is illustrated by eventual signs of left ventricular hypertrophy44,66,105 (3 cases), repolarization abnormalities in 7% of our series and in 10% in the literature9,34,74,78,81,144, and abnormal Q wave in 5%4,9,34,61,103,111.
All these electrical abnormalities reflect the granulomatous infiltration or its fibrous consequences on the conduction tissue and ventricular walls. The right bundle branch blocks and atrial and ventricular ectopic beats are significantly more frequent in patients with sarcoidosis than in a normal group of the same age and sex86. The prognostic implications of electrical abnormalities are differently appreciated according to the series, with seriousness of ventricular arrhythmias in the series of Sekiguchi et al105 versus atrioventricular blocks in ours. Bundle branch blocks have no influence on prognosis.
The importance of 24-hour Holter monitoring is variously appreciated in the literature. In our series, it was performed in 14% of cases and led to demonstration of other electrical abnormalities in 2 cases and more severe heart block in another case. The studies of Silverman et al114, Fleming36, Kinney et al61, and Suzuki et al119 plead in favor of systematic application in cases of CS suspicion. For Suzuki et al119, 24-hour Holter monitoring detected CS with a sensitivity of 67% and a specificity of 62%, when ventricular ectopic beats were more than 100 per day. If the ventricular ectopic beats were in group IV of Lown's classification, the sensitivity was similar but specificity was 80%.
In cases of conduction disturbance and/or severe arrhythmia139, or of unresponsiveness to cardiac or immunosuppressive therapy10,49,60,139, electrophysiologic studies are desirable, associated with a ventricular stimulation program. This examination, performed in 13 of our patients, was abnormal in 85% of cases. It was normal in 2 other patients studied for either severe night-bradycardia or severe atrial bigeminal rhythm (1 case each). The presence of late ventricular potential was noted twice out of 4. These patients had an abnormal rest-ECG, associated with a change of echocardiographic systolic or diastolic function. The indications of this examination are still not yet defined and are probably identical to those accepted for other cardiomyopathies.
The CS investigation is above all based on the ECG, which must be systematic in all patients with sarcoidosis, whether in the initial investigation of the disease or in follow-up phases. An electrical abnormality, even "insignificant," such as common sinus tachycardia (after eliminating any other origin, notably hyperthyroidism), must lead to investigation as it traces a myocardial infiltration86,99,103,114.
Echocardiography, together with isotopic studies, has transformed the diagnostic approach and surveillance of CS14,32,61,69,115,124. Performed in 40 of our patients, Doppler echocardiography was pathologic in 67% of cases. Observed abnormalities illustrated those described in the literature14,34,61,69,124,129, such as predominant septal and left ventricular abnormalities (see Table 5): dilation and/or dyskinesia of left ventricular wall. In our patients, pericardial involvement was evoked considering abnormal echoes (2 cases) and/or pericardial effusion (4 cases). A ventricular aneurysm was detected in 1 patient45,130.
Doppler echocardiography can also disclose miscellaneous findings, such as ventricular thrombus or mitral annular calcification61. Valvular involvement masquerading as metastatic tumor was recently reported56. A comparative study of CS and idiopathic dilated cardiomyopathy performed by Yazaki et al145 showed, in the CS group, a more frequent reduction of the anterior and apical segment of regional wall motion. The association of normokinetic, dyskinetic, or akinetic segments, due to the patchy distribution of the lesions, is suggestive of CS. Several authors2,14,32 insist on the importance of Doppler for researching ventricular diastolic dysfunction. This abnormality, noted in 29% of our patients, is more significantly observed in patients with CS than in a pulmonary sarcoidosis group32. It can be present even when ECGs and especially echocardiographies are normal, and seems to be connected to an abnormality of the active relaxation and not to a decrease of the left ventricular compliance32. The mechanisms are not clear: 1) an interstitial infiltration too small to be detected by echocardiography; 2) coronary microangiopathy; 3) associated factors, such as age or a higher systolic blood pressure, but these elements were not considered to be significant. All authors2,14,32 insist on the relevance of the Doppler study of the diastolic function, which, when altered, is a credible premature sign of a granulomatous CS. Echocardiographic abnormalities can be combined, as noted in 15 of our patients, and are not always associated with electrical signs14,32.
These echocardiographic abnormalities are all the more valuable when they are observed in young patients, without any other potential causes of heart disease.
During CS, there is a microvascular vasoconstriction, either in the active stage by local release of vasoactive mediators (histamine and/or serotonin, prostaglandin D2, leukotriene C4 and D4), or by regional metabolic abnormalities. Therefore cardiac isotopic scintigraphy is particularly instructive23,48,65,126,135, although a defect is not sufficient for diagnosing cardiac involvement. Results must be compared with clinical and laboratory data.
The most frequently used perfusion agents are thallium-201 and technetium-99m sestamibi, the latter being more sensitive67,77. Planar acquisition must be compared with that obtained after exercise imaging or after an infusion of dipyridamole with imaging during the redistribution phase. Improvement or complete resolution under dipyridamole or lack of worsening during effort allows differentiation of CS from proximal coronary arterial stenosis125,126,135. In the current series, radionuclide exams were performed in 40 patients, mainly with dipyridamole or exercise test. Three-quarters of the tests presented defects, present at least in septal or anteroseptal locations in 50% of cases. Defects were observed in all symptomatic cardiomyopathies, whereas they were not systematically present in patients with isolated arrhythmias or conduction abnormalities42. These defects can improve or resolve under corticosteroid therapy23,33,113,125 and reappear during relapses33; they can remain stable in cases of fibrosis. These different courses were observed in our patients during follow-up. Improvement of the defect after dipyridamole infusion is positively predictive for response to corticosteroids77.
On the whole, radionuclide imaging is justified when CS is suspected. Most studies show a good but inconstant42 correlation among 1) scintigraphic and echocardiographic abnormalities61 and 2) clinical and histologic studies (surgery, autopsies) and various scintigraphies: Gallium-6737,48,89,125, thallium-201, and technetium48. Diagnosis sensitivity of CS increases in the case of simultaneous abnormalities of gallium and thallium imaging37,48,77,123. The results of technetium scintigraphy are well correlated with those of echocardiography, notably for appreciation of the left ventricular function37,69. The 2 more specific perfusion agents, indium 111-labelled antimyosin antibody63 and MIBG84, have not clearly shown their superiority.
Magnetic Resonance Imaging
The first MRI studies17,27,29,98 are instructive and show, during CS, multiple high-signal lesions in T1 weighted sequences, rising in T2 acquisition and after gadolinium injection109. MRI allows for more precise myocardial biopsies, which can lead to a definitive diagnosis27,29,58,115. Complete resolution of MRI changes27,29,31,109,133 can be induced by corticosteroid treatment. Promising results reported with SPECT scan using technetium-9930 or thallium-201142 need to be confirmed by further studies.
There are at present no formal diagnostic criteria for CS, only elements of assumptions, as proposed by Yoshida et al147 in 1997 (Table 9). These criteria exclude mild to moderate forms that could be confirmed with laboratory signs other than those found in an ECG. It would be possible to differentiate a Group 2A with electrical signs and a Group 2B with laboratory signs which, together with relevant clinical signs, would include 2 or more of the items from b to f as listed in the revised criteria in Table 10, f standing for cardiac MRI abnormality. The ideal situation associating a suggestive clinical picture of CS, evolving within systemic sarcoidosis, associated with electrical, echocardiographic, scintigraphic abnormalities and the presence of cardiac granulomas, rarely exists or occurs too late (autopsy). The results of cardiac investigations consequently must be interpreted according to the context in which they evolve:
When cardiac signs are concomitant with systemic demonstrations of sarcoidosis, diagnosis is easy.
When the heart involvement appears in the months or years following the systemic signs, it is also justifiable to connect the heart signs with sarcoidosis, if all other causes of heart disease can be eliminated (see Table 9).
When cardiac signs are the presenting feature, there is then a diagnosis of probability, which at times can be confirmed by biopsies in asymptomatic organs (such as bronchial, salivary gland, and liver biopsies) or confirmed later, due to the appearance of extracardiac manifestations.
Finally, in rare cases42, heart disorders remain totally isolated. Endomyocardial biopsies are regarded a must, despite their limited sensitivity and keeping in mind as mentioned above that results are not fully specific, due to the possible occurrence of granulomas in other diseases (Table 11).
In the current series, the diagnosis of CS was not always easy, although 95% of our patients underwent ECG, echocardiography, and radionuclide imaging. Of these 39 patients, only 28% had 3 abnormal examinations. However, the observed abnormalities and especially their evolution under treatment were such that it was difficult to question the diagnosis of CS. A small granulomatous infiltration can provoke conductive disturbances and/or arrhythmias, localized defect, and/or abnormal coronary microcirculation. Our series illustrates perfectly the apparent discrepancies between the simultaneous investigations of electrical activity, morphology, ventricular function, and microvascularization. Thus, a normal ECG does not eliminate heart involvement that can be expressed only by an echocardiographic abnormality (such as pericardial effusion, localized hypokinesia, abnormal relaxation) or a defect on radionuclide imaging (see Table 7). Conversely, an abnormal ECG, especially conduction disturbance or arrhythmia, can be the sole expression of CS. Radionuclide imaging and echocardiography were simultaneously abnormal in only 53% of cases (see Table 7). It might have been expected that the symptomatic cardiomyopathy would be accompanied by more laboratory abnormalities. In fact, only 6 patients (35%) had 3 abnormal examinations (see Figure 2). Paradoxically, 6 of the 24 asymptomatic patients (25%) had 3 abnormal examinations (see Figures 3 and 4).
Beside diagnostic difficulties for correlating cardiac signs with sarcoidosis, cardiac involvement is difficult to eliminate during systemic sarcoidosis. Apart from economic considerations, should these 3 examinations be performed in all patients with sarcoidosis? No study provides the answer. Based on our experience, we suggest it is useless during nonsevere sarcoidosis (articular, Löfgren's syndrome, cutaneous, salivary gland), useful during severe sarcoidosis (severe general signs, active lung involvement, multisystemic granulomatous process), and mandatory in the presence of peculiar localizations (neurologic, bone, nasal).
CS is responsible for 50% of deaths in patients with sarcoidosis52, mainly by sudden death (23%-68% of cases)36,99,114 and congestive cardiomyopathy (25% of cases)9,36,114. It is generally agreed that CS should be treated9,38,42,44,66,71,99,103,130. In our study, although most cases could be considered relatively mild forms of the disease, all patients but 2 (1 asymptomatic pericardial effusion and 1 symptomatic complete heart block leading to pacemaker implantation) were treated with corticosteroids, associated with immunosuppressive agents in 13 cases (32%).
The long follow-up (average 58 mo) of 39 of our patients allowed for an appreciation of the course of systemic sarcoidosis and CS, the possibilities of relapse, and the drug-related side effects. All extracardiac involvement was cured, leaving neurologic residual deficits in only 3 patients.
The evolution of CS was favorable in 90% of cases, with normalization or improvement of clinical and/or laboratory parameters. Stabilization was observed in 2 cases, and 2 patients worsened. These unsatisfactory courses were observed in 4 patients with symptomatic cardiomyopathy (2 patients in NYHA stage II, 1 in stage III, 1 in stage IV).
In the long term, 6 patients died. However, only 1 death, occurring in the postoperative period after heart transplantation, could be connected with CS. That patient was the only case tardily treated. The 5 other deaths were not of cardiac origin and presented with a relatively mild CS. They occurred after a mean follow-up of 33 months after diagnosis and treatment of CS, which was considered as cured at that time. Two patients, presenting no evidence of neurosarcoidosis, committed suicide, which pleads in favor of a psychologic approach to the management of long-term disorders.
Although sudden death has been reported frequently in CS1,34,43,95,99,114,131, it was not observed in the current study. This may be explained by the moderate forms, the small number of ventricular arrhythmias in our series, an early and long-lasting therapeutic attitude, and the "aggressive" treatment administered in 13 severe cases, with association of corticosteroids and immunosuppressive agents.
Despite the cardiac expression, it is generally accepted that cardiac clinical signs carry a bad prognosis in CS. We considered that even relatively mild signs reflect a significant severity of the disease with quite an impact on prognosis. Relatively mild signs were typical in our series; we observed a favorable evolution in 87% of cases. Twenty-one patients (54%) were cured without residual cardiac signs. An improvement was noted in 13 patients, with residual signs such as moderate dyspnea (NYHA stage II), or laboratory abnormalities devoid of clinical consequences. Three patients were stabilized while maintaining their quality of life, and 2 others worsened (1 was treated too late and the other remained untreated).
As concerns cardiac manifestations, a favorable course, including cure or improvement, was observed in more than 80% of cases. However, the outcome depended on the initial features of CS (see Figure 4). A total cure without residual signs was mainly observed in patients with conduction disturbances (75% of cases) and echocardiographic and/or scintigraphic abnormalities (66% of cases). Arrhythmias and cardiomyopathies shared, in our series, a less favorable prognosis, with a percentage of cure of 55% and 47%, respectively. Our results differ from data found in the literature. That might be due to the small number of patients and to few with ventricular tachycardia. In these 2 groups, some patients worsened under treatment.
CS relapsed in 9 patients with a clinical or laboratory expression similar to the initial presentation. These patients had no frank distinguishing features. They had received corticosteroids for an average of 42 months before relapse (and immunosuppressive agents in 2 cases). For 3 patients, the relapse appeared after steroid withdrawal. Three patients were cured following resumption of corticosteroids. In 5 cases, additional immunosuppressive agents were required to obtain 4 cures and 1 dramatic improvement. For 1 patient, the survey was too short to assess outcome.
The Japanese series105,144 insisted on the circumstances of death from CS, thus confirming its severity, at least in this ethnic group. The retrospective study by Yazaki et al144 showed that, for patients with stable ventricular function, 20% died of arrhythmias or of sudden death, 36% of refractory cardiac insufficiency, and in 44% of cases the cause was progressive congestive cardiac insufficiency. Prognostic factors have been identified from these retrospective studies and from prospective studies120,144. In old autopsy series99,114, the presence of clinical and/or electrical signs marked a massive myocardial infiltration. Conduction disturbances and arrhythmias seemed to have a better prognosis34,42,44,50,71,100,102,105,119. In our series, even though atrioventricular blocks disappeared in most cases, their presence was associated with further relapses and an unfavorable course. Among patients with complete heart block treated by pacemaker, use of corticosteroids led to a significant improvement in life expectancy59,144: 47 months with steroids versus 25 months without144.
Active cardiomyopathy (with activity judged from isotopic examinations) usually has a favorable course42,50,70,112,124, as opposed to fibrotic forms, which resist treatments71,99,108,116. Compared with idiopathic cardiomyopathy, the prognosis of treated sarcoidosis cardiomyopathy remains severe, with a 5-year life expectancy of 64% in the former versus 37% in the latter120.
For Takada et al120, males have a better prognosis, as have those in classes I-II of the NYHA classification. In our series, neither sex, ethnic origin, NYHA class, presence of symptomatic cardiomyopathy, or echographic data influenced long-term prognosis. Relapses were more frequent, but not significant, among the oldest patients; white patients; and those suffering from atrial arrhythmia, atrioventricular block, or a defect detected by scintigraphy. These differences were also evident in the ethnic groups, the most severe or relapsing forms being observed preferentially in white patients.
Besides these clinical elements, Takada et al120 also insist on laboratory factors: diffuse defects during scintigraphy, dilation and abnormal left ventricular function evaluated by echocardiography, and initial resistance to treatment34.
On the whole, treatment becomes urgent as soon as CS is diagnosed. Modalities must be guided partially by the search for prognostic factors, always keeping in mind the risks of spontaneous evolution toward progressive cardiac insufficiency and of sudden death.
Treatment of CS includes steroids associated in some cases with immunosuppressive agents, and, when appropriate, specific cardiac measures.
Steroids and immunosuppressive treatment
Corticosteroids represent the first line of treatment and must be proposed once diagnosis is established9,34,42,44,50,66,71,99,105, 106,143. Their efficiency has been demonstrated both on clinical symptoms38,50,68,90,130 and on cardiac examinations42,49,68,96,112, notably conduction disturbances59,129, arrhythmias, and sometimes ventricular stimulating programs31,49. Echocardiography can normalize, as can MRI images27,29,31,109. Corticosteroids also allow a regression of isotopic abnormalities on gallium63,89, technetium145, or coupled gallium/thallium imaging50,125,129,130.
Although the efficiency of steroids is admitted by all, dosages and duration vary depending on the series, or even on a case by case analysis. Therefore, the following recommendations should be regarded merely as reflecting the opinions of a majority of authors, including our own.
For most patients with CS, the initial dose of 1 mg/kg per day of prednisone should be maintained during 6-8 weeks. A clinical and laboratory evaluation should then be performed before considering a progressive tapering of steroids. Steroid administration must be prolonged for several years, or even lifelong according to some authors105. The lowest possible dose should be established using disease activity parameters, to limit side effects and any decrease in quality of life. Attempts at withdrawal can be envisaged for certain patients, keeping in mind the risks of relapse of sarcoidosis, which occurred in 23% of our patients, and of sudden death, which can be due to a rough discontinuation of steroids105.
In severe symptomatic forms, associating electrical, echographic, and/or scintigraphic abnormalities, we consider that steroids can be initiated with a 3-day administration of intravenous pulses of 15 mg/kg methylprednisone42. Our experience with patients with neurosarcoidosis20 has led us to adopt an even more aggressive regimen in severe forms of CS, by proposing adjunction of immunosuppressive agents to steroids20,25,51. This attitude is supported by the positive results of our retrospective study, that is, the absence of sudden death and a percentage of cure and improvement exceeding 85%, although data are less brilliant in the case of symptomatic cardiomyopathy, which reflects major myocardial infiltration. We consider that the use of immunosuppressive agents is justified in the following conditions: 1) initial severity of CS, 2) failure of a high dose of steroids during the attack phase, 3) requirement of a high maintenance dose and, 4) relapse under steroid treatment. Choosing between cyclophosphamide, methotrexate, or cyclosporin, one must take into account their respective contraindications, but to date, the choice results more from local habits than from evidence-based data. Later in the course, in patients with advanced cardiac insufficiency and no other active localization, the decision to stop the treatment must be discussed, because corticosteroids are ineffective against fibrosis71,99,108,116. Prerequisites for steroid withdrawal imply converging arguments suggesting the absence of disease activity, attested by radionuclide or MRI imaging for cardiac involvement133, and by serum determinations of angiotensin converting enzyme in the absence of therapy with angiotensin converting enzyme inhibitors, lysozyme, or even IL1040.
Symptomatic treatments used in CS are variable10,87: digitalo-diuretics42,90, beta blockers, antiarrhythmics42,90, angiotensin converting enzyme inhibitors42,82, and angiotensin II receptor antagonists42,90. Anticoagulants seem to be justified in the presence of arrhythmias, dilated cardiomyopathy, and ventricular aneurysm (2 cerebral embolism events were observed in our series). When ventricular arrhythmias cannot be controlled medically, an automated implantable cardiac defibrillator may be indicated8,10,16,46,83,93,120,138,139. In all cases, a defibrillator was implanted after antiarrhythmics failed and after induced ventricular tachycardia had been considered, whether upon a stimulation test or after cardiac failure8. In a few cases, a pacemaker implantation did not prevent proceeding with heart transplantation93. In selected cases, a surgical resection10,138 or an ablation with radiofrequency catheter of the pathologic ventricular areas24,128 has been proposed. Pacemaker implantation remains the treatment of choice for advanced atrioventricular block, always in association with corticosteroids105,120. In rare cases, pericardectomy51 or ventricular aneurysm resection have been performed45.
Finally, a few heart transplantations were performed in patients with cardiomyopathy resistant to all therapeutic agents42,57,82,83,91,93,102,107,130. Indications remain imprecise and should be decided on a case-by-case basis. To date, only 1 patient suffered from a recurrence of the granulomatous disease on the transplant, which improved following an increase of corticosteroids91.
Several conclusions can be drawn from this retrospective study of 41 cases of CS, mainly affecting white patients, who were followed for a mean period of 58 months. Despite the male predominance, the mean age of patients in our series was high. CS arises preferentially during systemic sarcoidosis, and especially in the follow-up, during steroid tapering. The search for cardiac involvement must be systematic in any active sarcoidosis (clinical examinations and ECG), especially in the presence of neurologic manifestations. Repeating this search during the course of the disease leads to diagnosis in 85% of cases. Our series confirms the predominance of conduction disturbances and cardiomyopathy, and highlights the prevalence of atrial arrhythmias. Echocardiography and isotopic examinations, and cardiac MRI when possible, are mandatory in the evaluation of CS, to appreciate both its severity and activity. They must be systematically performed when sarcoidosis is characterized by a rare localization (neurologic, renal, bone, laryngeal) or when confronted with a progressive, severe, and multiorgan sarcoidosis. Beside the appropriate cardiac treatment, we used corticosteroids associated with immunosuppressive agents in 13 patients. This regimen allowed an improvement in 87% of cases, and, in particular, no sudden death occurred. Corticosteroids should be started at a high dosage (1 mg/kg per day) and associated with immunosuppressive agents in severe, resistant, or relapsing forms. Steroid administration must be prolonged for years, if not lifelong.
1. Abeler V. Sarcoidosis of the cardiac conducting system. Am Heart J
2. Angomachalelis N, Hourzamanis A, Vamvalis C, Gavrielides A. Doppler echocardiographic evaluation of left ventricular diastolic function in patients with systemic sarcoidosis. Postgrad Med J
3. Angomachalelis N, Hourzamanis A, Salem N, Vakalis D, Serasli E, Efthimiadis T, Triantaphyllon I. Pericardial effusion concomitant with specific heart muscle disease in systemic sarcoidosis. Postgrad Med J
. 1994;70(Suppl 1):S8-S12.
4. Anzai H, Momiyama Y, Kimura M. Cardiac sarcoidosis complicated by multivessel coronary spasm: a case report. J Cardiol
5. Arunabh S, Verma N, Brady TM. Massive pericardial effusion in sarcoidosis. Am Fam Physician
6. Auger G, Michenet P, Cami M, Kerdraon R. [Sudden death in an adult and detection of a cardiac site sarcoidosis.] Ann Med Interne (Paris)
7. Badorff C, Schwimmbeck PL, Kuhl U, Gerhold M, Stein H, Schultheiss HP. [Cardiac sarcoidosis: diagnostic validation by endomyocardial biopsy and therapy with corticosteroids.] Z Kardiol
8. Bajaj AK, Kopelman HA, Echt DS. Cardiac sarcoidosis with sudden death: treatment with automatic implantable cardioverter defibrillator. Am Heart J
9. Bashour FA, McConnell T, Skinner W, Manson M. Myocardial sarcoidosis disease. Chest
10. Becker D, Berger E, Chmielewski C. Cardiac sarcoidosis: a report of four cases with ventricular tachycardia. J Cardiovasc Electrophysiol
11. Bertola G, Torchio G, Meregalli M, Sironi C, Benvenuto M. [Heart sarcoidosis. A case of mitral and ascending aorta involvement.] Recenti Prog Med
12. Blanc JJ, Guillerm D, Boschat J, Clavier J, Kerbrat G, Penther PH. [Sinus node function and atrioventricular conduction in 20 cases of sarcoidosis.] Arch Mal Coeur Vaiss
13. Bohle W, Schaefer HE. Predominant myocardial sarcoidosis. Pathol Res Pract
14. Burstow DJ, Tajik AJ, Bailey KR, De Remee RA, Taliercio CP. Two-dimensional echocardiographic findings in systemic sarcoidosis. Am J Cardiol
15. Case Records of the Massachusetts General Hospital (case 34-1985). N Engl J Med
16. Case Records of the Massachusetts General Hospital (case 16-1995). N Engl J Med
17. Chandra M, Silverman ME, Oshinski J, Pettigrew R. Diagnosis of cardiac sarcoidosis aided by MRI. Chest
18. Chapelon C, Uzzan B, Piette JC, Jacques C, Coche C, Godeau P. [Sarcoidosis in internal medicine. A cooperative study of 554 cases.] Ann Med Interne (Paris)
19. Chapelon C, Piette JC, Uzzan B, Coche E, Herson S, Ziza JM, Godeau P. [The advantages of histological samples in sarcoidosis. Retrospective multicenter analysis of 618 biopsies performed on 416 patients.] Rev Med Interne
20. Chapelon C, Ziza JM, Piette JC, Levy Y, Raguin G, Wechsler B, Bitker MO, Bletry O, Laplane D, Bousser MG, Godeau P. Neurosarcoidosis: signs, course and treatment in 35 confirmed cases. Medicine (Baltimore)
21. Chapelon-Abric C, Lascault G, De Gennes C, Godeau P. [Cardiac sarcoidosis. A case of sinoatrial block.] Presse Med
22. Chapelon-Abric C. [Cardiac sarcoidosis.] Ann Med Interne (Paris)
23. Chin BB, Civelek AC, Mudun A. Resting TI-201 scintigraphy in the evaluation of myocardial sarcoidosis. Clin Nuclear Med
24. Delacretaz E, Stevenson WG, Winters GL. Ablation of ventricular tachycardia with a saline-cooled radiofrequency catheter: anatomic and histologic characteristics of the lesions in humans. J Cardiovasc Electrophysiol
25. Demeter SL. Myocardial sarcoidosis unresponsive to steroids. Treatment with cyclophosphamide. Chest
26. Diderholm E, Eklund A, Orinius E, Widstrom O. Exudative pericarditis in sarcoidosis. A case report and echocardiographic study. Sarcoidosis
27. Doherty MJ, Kumar SK, Nicholson AA, McGivern DV. Cardiac sarcoidosis: the value of magnetic resonance imaging in diagnosis and assessement of response to treatment. Respir Med
28. Drui S, Brandt CM, Neiman JL, Fincker JL. [Cardiac sarcoidosis. Mechanisms of valvular dysfunctions.] Sem Hop
29. Dupuis JM, Victor J, Furber A, Pezard P, Lejeune LL, Tadei A. [Value of magnetic resonance imaging in cardiac sarcoidosis. Apropos of a case.] Arch Mal Coeur Vaiss
30. Eguchi M, Tsuchihashi K, Hotta D, Hashimoto A, Sasao H, Yuda S, Nakata T, Shijubou N, Abe S, Shimamoto K. Technetium-99m sestamibi/tetrofosmin myocardial perfusion scanning in cardiac and noncardiac sarcoidosis. Cardiology
31. Eliasch H, Juhlin-Dannfelt A, Sjogren I, Terent A. Magnetic resonance imaging as an aid to the diagnosis and evaluation of suspected myocardial sarcoidosis in a fighter pilot. Aviat Space Environ Med
32. Fahy GJ, Marwick T, McCreery CJ, Quigley PJ, Maurer BJ. Doppler echocardiographic detection of left ventricular diastolic dysfunction in patients with pulmonary sarcoidosis. Chest
33. Fields CL, Ossorio MA, Roy TM, Denny DM, Varga DW. Thallium-201 scintigraphy in the diagnosis and management of myocardial sarcoidosis. South Med J
34. Fleming HA. Sarcoid heart disease. Br Heart J
35. Fleming HA. Sarcoid heart disease. Br Heart J
36. Fleming HA. Death from sarcoid heart disease. United Kingdom series 1971-1986. 300 cases with 138 deaths. In: Grassi G. Rizzato G, Pozzi E, eds. Sarcoidosis and Other Granulomatous Disorders: Proceedings of the 11th World Congress on Sarcoidosis and Other Granulomatous Disorders
. Amsterdam: Elsevier; 1988:19-33.
37. Forman MB, Sandler MP, Sacks GA, Kronenberg MW, Powers TA. Radionuclide imaging in myocardial sarcoidosis. Demonstration of myocardial uptake of technetium pyrophosphate99m and gallium. Chest
38. Fujita N, Hiroe M, Suzuki Y, Sato H, Inoue Y, Sekiguchi M, Hosoda S. A case with cardiac sarcoidosis. Significance of the effect of steroids on the reversion of advanced atrioventricular block and myocardial scintigraphic abnormalities. Heart Vessels Suppl
39. Fukuhara T, Morini M, Sakoda S, Bito K, Kinoshita M, Kawakita S. Myocarditis with multinucleated giant cells detected in biopsy specimens. Clin Cardiol
40. Fuse K, Kodama M, Okura Y, Ito M, Aoki Y, Hirono S, Kato K, Hanawa H, Aizawa Y. Levels of serum interleukin-10 reflect disease activity in patients with cardiac sarcoidosis. Jpn Circ J
41. Gibbons WJ, Levy RD, Nava S, Malcolm J, Marin JM, Tardif C, Magder C, Lisbona R, Cosio MG. Subclinical cardiac dysfunction in sarcoidosis. Chest
42. Godeau P, Chapelon-Abric C. [Sarcoidosis symptomatic cardiomyopathy: report of 10 cases.] Bull Acad Natl Med
43. Gori F, Taddei GL, Pedemonte E, Ciaccheri M. [Sarcoidosis with predominant cardiac involvement.] Cardiologica
44. Gozo EG, Cosnow I, Cohen HC, Okun L. The heart in sarcoidosis. Chest
45. Grollier G, Galateau F, Scanu P, Commeau P, Voglimacci M, Bernard JP, Khayat A, Potier JC. [Cardiac sarcoidosis responsible for localized left ventricular ectasia and refractory ventricular tachycardia. Anatomoclinical study.] Arch Mal Coeur Vaiss
46. Hermosilla Cabrerizo T, Rodrigo Trallero G, Suarez Alzamora J, Pelegrin Diaz J, Zabala Lopez S, Diez Manglano J, Bueno Gmez J. [Myocardial sarcoidosis treated with implantable defibrillator.] An Med Interna
47. Hillerdal O, Niou E, Osterman K. Sarcoidosis. Epidemiology and prognosis. A 15-year European study. Am Rev Respir Dis
48. Hirose Y, Ishida Y, Hayashida K, Maeno M, Takamiya M, Ohmori F, Miyatake K, Uehara T, Nishimura T, Tachibana T. Myocardial involvement in patients with sarcoidosis. An analysis of 75 patients. Clin Nucl Med
49. Huang PL, Brooks R, Carpenter CH, Garan H. Antiarrhythmic therapy guided by programmed electrical stimulation in cardiac sarcoidosis with ventricular tachycardia. Am Heart J
50. Ishikawa T, Kondoh H, Nakagawa S, Koiwaya Y, Tanaka K. Steroid therapy in sarcoidosis. Increased left ventricular contractility concomitant with electrocardiographic improvement after prednisone. Chest
51. Israel RH, Poe RH. Massive pericardial effusion in sarcoidosis. Respiration
52. Iwai K, Tachibana T, Takemura T, Matsui Y, Kitaichi M, Kawabata Y. Pathological studies on sarcoidosis autopsy. I. Epidemiological features of 320 cases in Japan. Acta Pathol Jpn
53. Iwai K, Sekiguti M, Hosoda Y, DeRemee RA, Tazelaar HD, Sharma OP, Maheshwari A, Noguchi TI. Racial difference in cardiac sarcoidosis incidence observed at autopsy. Sarcoidosis
54. Jain A, Starek PJ, Delany DL. Ventricular tachycardia and ventricular aneurysm due to unrecognized sarcoidosis. Clin Cardiol
55. James DJ, Turiaf J, Hosoda YY. Description of sarcoidosis: report of the subcommittee on classification and definition. In: Siltzbach LE, ed. Seventh International Conference of Sarcoidosis and Other Ganulomatous Disorders
. New York: New York Academy of Sciences; 1976:742.
56. Joffe II, Lampert C, Jacobs LE, Owen AN, Ioli A, Kotler MN. Cardiac sarcoidosis masquerading as a metastatic tumor: the role of transthoracic and transesophageal echocardiography. J Am Soc Echocardiogr
57. Johns CJ, Paz H, Kasper EK, Baughman K. Myocardial sarcoidosis: course and management. Sarcoidosis
58. Kasai H, Suzuki J, Imamura H, Yazaki Y, Isobe M, Sekiguchi M. A case of cardiac sarcoidosis with advanced atrioventricular block. Failure of endomyocardial biopsy diagnosis and success in detecting. Heart Vessels
59. Kato Y, Morimoto S, Uemura A, Hiramitsu S, Ito T, Hishida H. Efficacy of corticosteroids in sarcoidosis presenting with atrioventricular block. Sarcoidosis Vasc Diffuse Lung Dis
60. Kawamura Y, Yoshida A, Toyoshima E, Sato N, Kawashima E, Hasebe N, Kikuchi K, Matsuhashi H. A case of cardiac sarcoidosis: significance of ventricular tachycardia originating from the septum. Jpn Circ J
61. Kinney EL, Jackson GL, Reeves WX, Zelis R. Thallium-scan myocardial defects and echocardiographic abnormalities in patients with sarcoidosis without clinical cardiac dysfunction. An analysis of 44 patients. Am J Med
62. Kinney E, Murthy R, Ascunge GL, Donohoe R. Pericardial effusion in sarcoidosis. Chest
63. Knapp WH, Bentrup A, Ohlmeier H. Indium 111-labelled antimyosin antibody imaging in a patient with sarcoidosis. Eur J Nucl Med
64. Kollermann J, Roos G, Helpap B. Sudden cardiac death from unrecognized cardiac sarcoidosis. Pathology
65. Kurata C, Sakata K, Taguchi T. SPECT imaging with TI-201 and Ga-67 in myocardial sarcoidosis. Clin Nucl Med
66. Lanfranchi J, Battesti JP, Habbas JP, d'Angeli JL. [Electrocardiographic anomalies in mediastino-pulmonary sarcoidosis without cor pulmonale.] Rev Med Interne
67. LeGuludec D, Menad F, Faraggi M, Weinmann P, Battesti JP, Valeyre D. Myocardial sarcoidosis: clinical value of technetium-99m sestamibi tomoscintigraphy. Chest
68. Lemery R, McGoon MD, Edwards WD. Cardiac sarcoidosis: a potentially treatable form of myocarditis. Mayo Clin Proc
69. Lewin RF. Echocardiographic evaluation of patients with systemic sarcoidosis. Am Heart J
70. Lip GYH, Gupta J, Gill JS, Singh SP. Sarcoid heart disease: a rare cause of chest pain and malignant cardiac arrhythmia in a young Asian man. Angiology
71. Lorell B, Alderman EL, Mason JW. Cardiac sarcoidosis: diagnosis with endomyocardial biopsy and treatment with corticosteroids. Am J Cardiol
72. Lower EE, Baughman RP. Prolonged use of methotrexate for sarcoidosis. Arch Intern Med
73. Ludmerer KM, Kissane JM. Refractory ventricular arrhythmias and death in a 43-year old man. Am J Med
74. 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 Cardiol
75. Mafart B, Talard P, Bouchiat C, Chagnon A. [Does corticosteroid therapy favor the development of ventricular aneurysm in myocardial sarcoidosis?] Arch Mal Coeur Vaiss
76. Mana J. Nuclear imaging. 67Gallium, 201thallium, 18F-labeled fluoro-2-deoxy-D-glucose positron emission tomography. Clin Chest Med
77. Matsui Y, Iwai K, Tachibana T, Fruie T, Shigematsu N, Izumi T, Homma AH, Mikami R, Hongo O, Hiraga Y, Yamamoto M. Clinicopathological study of fatal myocardial sarcoidosis. Ann N Y Acad Sci
78. Matsumori A, Hara M, Nagai S, Izumi T, Ohashi N, Ono K, Sasayama S. Hypertrophic cardiomyopathy as a manifestation of cardiac sarcoidosis. Jpn Circ J
79. Mayock RL, Bertrand DP, Morrison CE, Scott JH. Manifestations of sarcoidosis. Analysis of 145 patients with a review of nine series selected from the literature. Am J Med
80. Mazzone P, Arroliga AC. Acute dyspnea and hypoxia in a 37-year-old woman with sarcoidosis. Chest
81. McDougall NI, Purvis JA, Wilson CM, Adgey AA. Asystolic arrest as a presentation of sarcoidosis. Int J Cardiol
82. Messas E, Charniot JC, Lanfranchi J, Battesti JP, Weinman P, Zerhouni K, Gallot D, Artigou JY. [Cardiogenic shock presenting a cardiac sarcoidosis. A propos of a case.] Arch Mal Coeur Vaiss
83. Meyer A, Schafer H, Doring V, Maisch B, Kirsten D. [Heart transplantation in myocardial sarcoidosis. Studies on the explanted heart.] Dtsch Med Wochenschr
84. Misumi I, Kimura Y, Hokamura Y, Honda Y, Yasunaga T, Nakashima K, Takemura N, Asoshina M, Uranaka N, Takenaka S, Shima K. Scintigraphic detection of regional disruption of adrenergic nervous system in sarcoid heart disease. Jpn Circ J
85. Naruse TK, Matsuzawa Y, Ota M, Katsuyama Y, Matsumori A, Hara M, Nagai S, Morimoto S, Sasayama S, Inoko H. HLA-DQB1*0601 is primarily associated with the susceptibility to cardiac sarcoidosis. Tissue Antigens
86. Numao Y, Sekiguchi M, Furuie T, Matsui Y, Izumi T, Mikami R. A study of cardiac involvement in 963 cases of sarcoidosis by ECG and endomyocardial biopsy. In: Jones Williams B, Davies BH, eds. 8th International Conference on Sarcoidosis and Other Granulomatous Diseases. 1978
. Cardiff, Wales: Alpha Omega Publishing; 1980:607-614.
87. Okamoto M, Hashimoto M, Sueda T, Munemori M, Yamada T. Polymorphic ventricular tachycardia with cardiac sarcoidosis: treatment with low dose metoprolol and cibenzoline. Intern Med
88. Okamoto H, Mizuno K, Ohtoshi E. Cutaneous sarcoidosis with cardiac involvement. Eur J Dermatol
89. Okayama K, Kurata C, Tawarahara K, Wakabayashi Y, Chida K, Sato A. Diagnostic and prognostic value of myocardial scintigraphy with thallium-201 and gallium-67 in cardiac sarcoidosis. Chest
90. Oldenburg O, Schafers RF, Kuntz S, Sack S, Erbel R, Philipp T, Weber F. [Heart failure as a cardiac symptom of sarcoidosis. Successful treatment of heart failure with steroids, digitalis and an angiotensin-1-receptor in sarcoidosis.] Med Klin (Munich)
91. Oni AA, Hershberger RE, Norman DJ, Ray J, Hovaguimian H, Cobanoglu AM, Hosenpud JD. Recurrence of sarcoidosis in a cardiac allograft: control with augmented corticosteroids. J Heart Lung Transplant
92. Ott P, Marcus FI, Sobonya F, Knight BP, Fuenzalida CP. Cardiac sarcoidosis masquerading as right ventricular dysplasia. Pacing Clin Electrophysiol
93. Paz HL, McCormick DJ, Kutalek SP, Patchefsky A. The automated implantable cardiac defibrillator. Prophylaxis in cardiac sarcoidosis. Chest
94. Perry A, Vuitch F. Causes of death in patients with sarcoidosis. A morphologic study of 38 autopsies with clinicopathologic correlations. Arch Pathol Lab Med
95. Porter GH. Sarcoid heart disease. N Engl J Med
96. Ratner SJ, Fenoglio JJ, Ursell PC. Utility of endomyocardial biopsy in the diagnosis of cardiac sarcoidosis. Chest
97. Reuhl J, Schneider M, Sievert H, Lutz FU, Zieger G. Myocardial sarcoidosis as a rare cause of sudden cardiac death. Forensic Sci Int
98. Riedy K, Fisher MR, Belic N, Koenigsberg DI. MR imaging of myocardial sarcoidosis. AJR Am J Roentgenol
99. Roberts WC, McAllister HA, Ferrans VJ. Sarcoidosis of the heart: a clinicopathologic study of 35 necropsy patients and review of 78 previously described necropsy patients. Am J Med
100. Rubinstein I, Baum GL, Hiss Y. Cardiac tamponade as the presenting symptom of sarcoidosis. Am Heart J
101. Rybicki BA, Major M, Popovich J Jr, Maliarik MJ, Iannuzzi MC. Racial differences in sarcoidosis incidence: a 5-year study in a health maintenance organization. Am J Epidemiol
102. Schaedel H, Kirsten D, Schmidt H, Strauss HJ. Sarcoid heart disease: results of follow-up investigations. Eur Heart J
103. Sekiguchi M, Numao Y, Imai M, Furuie T, Mikami R. Clinical and histopathological profile of sarcoidosis of the heart and acute idiopathic myocarditis. Concepts through a study employing endomyocardial biopsy. II. Myocarditis. Jpn Circ J
104. Sekiguchi M, Kaneko M, Hiroe M, Hirosawa K. Recent trends in cardiac sarcoidosis research in Japan. Heart Vessels Suppl
105. Sekiguchi M, Yazaki Y, Isobe M, Hiroe M. Cardiac sarcoidosis: diagnostic, prognostic and therapeutic considerations. Cardiovasc Drugs Ther
106. Shammas RL, Movahed A. Successful treatment of myocardial sarcoidosis with steroids. Sarcoidosis
107. Sharma OP, Klatt E. Factors which adversely affect the course of sarcoidosis. In: Grassi G, Rizzato G, Pozzi E, eds. Sarcoidosis and Other Granulomatous Disorders: Proceedings of the 11th World Congress on Sarcoidosis and Other Granulomatous Disorders
. Amsterdam: Elsevier; 1988:421-422.
108. Shiff AD, Blatt CJ, Colp C. Recurrent pericardial effusion secondary to sarcoidosis of the pericardium: a biopsy proved case. N Engl J Med
109. Shimada T, Shimada K, Sakane T, Ochiai K, Tsukihashi H, Fukui M, Inoue S, Katoh H, Murakami Y, Ishibashi Y, Maruyama R. Diagnosis of cardiac sarcoidosis and evaluation of the effects of steroid therapy by gadolinium-DTPA-enhanced magnetic resonance imaging. Am J Med
110. Shimazu A, Ishikura Y, Odagiri S, Hashimoto M, Hirao D, Shirakusa T. [A case of mitral stenosis associated with cardiac sarcoidosis and left ventricular aneurysm.] Nippon Kyobu Geka Gakkai Zasshi
111. Shindo T, Kurihara H, Ohishi N, Morita H, Maemua K, Kurihara Y, Tsuneyoshi H, Chi H, Yamaski K, Yazaki Y. Cardiac sarcoidosis. Circulation
112. Shiotani H, Miyazaki T, Matsunaga M, Kado T. Improvement of severe heart failure with corticosteroid therapy in a patient with myocardial sarcoidosis. Jpn Circ J
113. Shiraishi J, Tatsumi T, Shimoo K, Katsume A, Mani H, Kobara M, Shirayama T, Azuma A, Nakagawa M. Cardiac sarcoidosis mimicking right ventricular dysplasia. Circ J
114. Silverman KJ, Hutchins GM, Bulkley BH. Cardiac sarcoid: A clinicopathologic study of 84 unselected patients with systemic sarcoidosis. Circulation
115. Skold CM, Larsen FF, Rasmussen E, Pehrsson SK, Eklund AG. Determination of cardiac involvement in sarcoidosis by magnetic resonance imaging and doppler echocardiography. J Intern Med
116. Stein E, Jackler I, Stimmel B. Asymptomatic electrocardiographic alterations in sarcoidosis. Am Heart J
117. Stewart RE, Graham DM, Godfrey GW, Friedman HZ, Das SK, Shea MJ. Rapidly progressive heart failure resulting from cardiac sarcoidosis. Am Heart J
118. Studdy RP, Lapworth R, Bird R. Angiotensin-converting enzyme and its clinical significance: a review. J Clin Pathol
119. Suzuki T, Kanda T, Kubota S, Imai S, Murata K. Holter monitoring as a noninvasive indicator of cardiac involvement in sarcoidosis. Chest
120. Takada K, Ina Y, Yamamoto M, Satoh T, Morishita M. Prognosis after pacemaker implantation in cardiac sarcoidosis in Japan. Clinical evaluation of corticosteroid therapy. Sarcoidosis
121. Takahashi T, Fukuta S, Date T, Hamamoto T, Yoshinaga T, Kohtoku S, Ono S, Matsuzaki M, Kusukawa R. An autopsy case of cardiac sarcoidosis with refractory ventricular tachycardia. Heart Vessels Suppl
122. Takashige N, Naruse TK, Matsumori A, Hara M, Nagai S, Morimoto S, Hiramitsu S, Sasayama S, Inoko H. Genetic polymorphisms at the tumor necrosis factor loci (TNFA and TNFB) in cardiac sarcoidosis. Tissue Antigens
123. Taki J, Nakajima K, Bunko H, Ohguchi M, Jonami N, Hisada K. Cardiac sarcoidosis demonstrated by T1-201 and Ga-67 SPECT imaging. Clin Nucl Med
124. Tan LB, Dickie S, McKenna WJ. Left ventricular diastolic characteristics of cardiac sarcoidosis. Am J Cardiol
125. Tawarahara K, Kurata C, Okayama K, Kobayashi A, Yamazaki N. Thallium-201 and gallium 67 single photon emission computed tomographic imaging in cardiac sarcoidosis. Am Heart J
126. Tellier P, Paycha F, Antony I, Nitenberg A, Valeyre D, Foult JM, Battesti JP. Reversibility by dipyridamole of thallium-201 myocardial scan defects in patients with sarcoidosis. Am J Med
127. Ueda M, Fujimoto T, Shoji S, Nagao M, Sorachi K, Ito T. Cardiac sarcoidosis. Jpn Heart J
128. Uemura A, Morimoto S, Hiramitsu S, Kato Y, Ito T, Hishida H. Histologic diagnostic rate of cardiac sarcoidosis: evaluation of endomyocardial biopsies. Am Heart J
129. Umetani K, Ishihara T, Yamamoto K, Sawanobori T, Kohno I, Ijiri H, Komori S, Tamura K. Successfully treated complete atrioventricular block with corticosteroid in a patient with cardiac sarcoidosis: usefulness of gallium-67 and thallium-201 scintigraphy. Intern Med
130. Valantine HA, Tazelaar HD, Macoviak J, Mullin AV, Hunt SA, Feuler MB, Bellingham ME, Schroeder JS. Cardiac sarcoidosis: response to steroids and transplantation. J Heart Transplant
131. Veinot JP, Johnston B. Cardiac sarcoidosis, an occult cause of sudden death: a case report and literature review. J Forensic Sci
132. Verkleeren JL, Glover MU, Bloor C, Joswig BC. Cardiac tamponade secondary to sarcoidosis. Am Heart J
133. Vignaux O, Dhote R, Duboc D, Blanche P, Dusser D, Weber S, Legmann P. Clinical significance of myocardial magnetic resonance abnormalities in patients with sarcoidosis: a 1-year follow-up study. Chest
134. Virmani R, Bures C, Roberts WC. Cardiac sarcoidosis. A major cause of sudden death in young individuals. Chest
135. Wait JL, Movahed A. Anginal chest pain in sarcoidosis. Thorax
136. Wallaert B, Ramon PH, Fournier E. Bronchoalveolar lavage, serum angiotensin converting enzyme and gallium 67 scanning in extra-thoracic sarcoidosis. Chest
137. Walsh MJ. Systemic sarcoidisis with refractory ventricular tachycardia and heart failure. Br Heart J
138. Watson J, Smith V, Schmidt D, Navratil D. Automatic implantable cardioverter-defibrillator: early experience at Wilford Hall USAF Medical Center. South Med J
139. Winters SL, Cohen M, Greenberg S, Stein B, Curwin J, Pe E, Gomes JA. Sustained ventricular tachycardia associated with sarcoidosis: assessment of the underlying cardiac anatomy and the prospective utility of programmed ventricular stimulation, drug therapy and an implantable antitachycardia device. J Am Coll Cardiol
140. Wynne JW, Ryerson GG, Dalovisio J. Myocardial sarcoidosis complicated by mural thrombosis. Thorax
141. Yamaguchi J, Shikano Y, Sato S, Shimamoto K. [A clinical and pathological study of 46 cases of sudden and unexpected death.] Nippon Ronen Igakkai Zasshi
142. Yamamoto N, Goth K, Yagi Y, Terashima Y, Nagashima K, Sawa T, Deguchi F, Nawada M, Tanaka H, Tsukamoto T. Thallium-201 myocardial SPECT findings at rest in sarcoidosis. Ann Nucl Med
143. Yamamoto M, Muramatsu M, Suzuki T. Successful corticosteroid treatment of seven cases of probable myocardial sarcoid. In: Jones Williams B, Davies BH, eds. International Conference on Sarcoidosis and Other Granulomatous Diseases
. Cardiff, Wales: Alpha Omega Publishing; 1980;615-623.
144. Yazaki Y, Hongo M, Hiroyoshi Y. Cardiac sarcoidosis in Japan: treatment and prognosis. In: Sekiguchi M, Ridchardson PJ, eds. Prognosis and Treatment of Cardiomyopathy and Myocarditis
. Tokyo: University of Tokyo Press; 1994:351-353.
145. Yazaki Y, Isobe M, Hayasaka M, Tanaka M, Fujii T, Sekiguchi M. Cardiac sarcoidosis mimicking hypertrophic cardiomyopathy: clinical utility of radionuclide imaging for differential diagnosis. Jpn Circ J
146. Yazaki Y, Isobe M, Hiramitsu S, Morimoto S, Hiroe M, Omichi C, Nakano T, Saeki M, Izumi T, Sekiguchi M. Comparison of clinical features and prognosis of cardiac sarcoidosis and idiopathic dilated cardiomyopathy. Am J Cardiol
147. Yoshida Y, Morimoto S, Hiramitsu S, Tsuboi N, Hirayama H, Itoh T. Incidence of cardiac sarcoidosis in Japanese patients with high-degree atrioventricular block. Am Heart J
148. Zelcer AA, LeJemtel TH, Jones J, Stahl J. Pericardial tamponade in sarcoidosis. Can J Cardiol