Intended for healthcare professionals

Editorials

Cardiac sarcoidosis

BMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7077.320 (Published 01 February 1997) Cite this as: BMJ 1997;314:320

A potentially fatal condition that needs expert assessment

  1. D N Mitchell, Honorary consultant physiciana,
  2. R M du Bois, Consultant respiratory physiciana,
  3. P J Oldershaw, Consultant cardiologista
  1. a Directorates of Respiratory Medicine and Cardiology, Royal Brompton Hospital, London SW3 6NP

    Cardiac arrhythmias and sudden death are likely among patients known to have cardiac lesions from sarcoidosis. Other clinical presentations include conduction defects, congestive heart failure, and a picture suggesting myocardial infarction–but the first evidence of cardiac sarcoidosis may be provided by a pathologist after the sudden death of someone who had been completely free of symptoms.1

    Attempts have been made to assess the prevalence of cardiac sarcoidosis. Mikhail et al studied 147 consecutive patients presenting with sarcoidosis and found 14 with electrocardiographic abnormalities; 10 of these had no symptoms related to the heart.2 Stein et al surveyed 80 selected patients aged under 40 years, of whom only five were white and 47 had chronic disease.3 Electrocardiographic abnormalities of conduction or repolarisation were detected in 41 of these 80 patients, none of whom had cardiac symptoms.

    In two other large series of unselected patients the proportion with clinical evidence of granulomatous changes in the heart was small. Scadding and Mitchell reported that, of 500 consecutive patients under their care, only four showed such evidence.1 In the series of 702 patients reported by Hagemann and Wurm 5% had clinical evidence of cardiac sarcoidosis, although cardiac lesions were found at necropsy in 15%.4

    In necropsy studies of sarcoidosis the proportion of patients with granulomas in the heart is always higher than the proportion known to have had cardiac sarcoidosis during life. Matsui et al studied 42 patients with myocardial sarcoidosis; in only five of these had sarcoidosis been diagnosed in life, and death had been sudden in 16.5 Roberts et al reviewed 113 necropsies of patients with cardiac sarcoidosis, in 89 of whom it had been the cause of death.6 The diagnosis of sarcoidosis had been made during life in only 24, and death had been sudden in 60. Fleming collected the records of patients throughout Britain in whom the diagnosis of sarcoidosis had been reached either during life or at necropsy.7 He accumulated a series of 138 fatalities among 300 patients with sarcoid heart disease. At presentation 73% had ventricular or supraventricular arrhythmia, 26% had complete heart block, and 24% had congestive cardiomyopathy. Sixty one per cent had right or left bundle branch block or partial heart block. Sudden death had occurred in 77 cases, and in 49 cardiac sarcoidosis had not been diagnosed previously.

    Thus, the heart may have extensive sarcoid lesions without clinical symptoms or evidence of disease elsewhere. Sudden death is the most common first manifestation of cardiac sarcoidosis, though some patients have antecedent dysrhythmias. Can cardiac lesions be diagnosed earlier? We suggest that 24 hour electrocardiographic monitoring and echocardiography is performed and generalised granulomatosis confirmed by tissue biopsy of affected organs or by the Kveim-Siltzbach test.

    If coronary heart disease can be excluded perfusion defects in thallium-201 myocardial scintigraphy may suggest sarcoidosis, particularly if they are absent or smaller during exercise or after dipyridamole infusion.8 9 Other scintigraphic investigations currently being studied include the combination of thallium-201 and gallium-67 imaging of the heart, especially with the use of single photon emission computed tomography (SPECT),10 Fab fragments of antimyosin antibodies labelled with indium-111,11 tomoscintigraphy with technetium-99m,12 and gated cardiac magnetic resonance imaging.

    Endomyocardial biopsy may yield sarcoid granulomas, but, because of the patchy distribution of the disease, a negative result does not exclude the diagnosis.13 Oakley considers that sarcoid infiltration is one of the few causes of focal left ventricular abnormality in the absence of disease of the major coronary arteries.14 The distribution is typically in the proximal portion of the left ventricle and upper septum, where it may produce akinesia or hypokinesia.

    Myocardial sarcoidosis is difficult to diagnose and has a treacherous course that may be fatal. Corticosteroids may lead to clinical improvement but should be reserved for patients in whom the diagnosis of cardiac sarcoidosis is reasonably certain. Minor electrocardiographic changes occurring alone without any supporting evidence of myocardial damage are not an indication for treatment. The dose should be high–of the order of 60 mg prednisolone daily or more–and treatment will need to be continued with a maintenance regimen to preserve any improvement gained.

    In intractable cases and in patients who find the side effects of corticosteroids intolerable an alternative is the use of pulsed intravenous methylprednisolone with a lower maintenance dose of oral prednisolone.15 The addition of hydroxychloroquine, methotrexate, or cyclophosphamide may be helpful.16 17 A permanent endocardial pacemaker should be used early in patients with unstable or complete heart block; corticosteroids are usually needed in patients who have pacemakers. Reports such as that of Fleming and Bailey18 suggest that treatment with corticosteroids results in a reduction in the numbers of premature beats and episodes of tachycardia with easier control of arrhythmias, and an improvement may also be seen in conduction defects. The danger of sudden death may also be diminished. Cardiac transplantation remains a possibility for younger patients with severe intractable heart failure,19 but sarcoidosis may recur in the cardiac allograft.20

    This review indicates the absence of any agreed strategy for the diagnosis, prognosis, and assessment of outcome measures in response to treatment. We believe that patients in whom the diagnosis of cardiac sarcoidosis is thought likely on clinical and electrocardiographic grounds should be further investigated with Holter monitoring and echocardiography reported by a cardiologist with the relevant experience. Expert cardiological advice is needed before embarking on any further cardiological investigations, including myocardial biopsy.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.