This patient displayed several unusual features of sarcoidosis, namely: the large, bilateral pleural effusions, myocardial and pericardial involvement, lack of systemic “markers” for the disease process, and lack of response to steroids. It is estimated that approximately 1 percent of patients with sarcoidosis have pleural involvement. The presence of either bilateral or massive pleural effusions are usually the subjects of case reports. Sarcoidosis may affect any area of the heart—pericardium, myocardium, or endocardium. Of these, the myocardium is most frequently involved. The exact incidence of cardiac involvement, however, is difficult to ascertain. Autopsy series suggest an incidence between 20 and 30 percent, although these figures are probably quite high for all patients with sarcoidosis. Pericardial involvement, by contrast, is unusual. Only 31 cases of pericardial granulomas (usually diagnosed at necropsy) were identified by 1979. The frequency of small pericardial effusions (which may be due to disease affecting the pericardium or the epicardium) appears to be subject to the diligence of the search. Nineteen percent of 48 consecutive sarcoidosis patients in a series by Kinney et al had small pericardial effusions diagnosed by echocardiography. Thus, it is unclear how often and to what extent sarcoidosis may affect the heart.

Corticosteroid medications are the usual mainstay of treatment for sarcoidosis. Cytotoxic medications have been used only rarely (mainly because steroids are almost uniformly effective, although an exact incidence could not be found in a literature search). Kataria used chlorambucil therapy in ten patients with sarcoidosis, either alone or in combination with steroids; eight improved. All ten were placed on therapy with this agent after a lack of medical response to steroids or contraindications to their use. Similarly, Israel et al used chlorambucil in eight patients not responding to corticosteroid therapy, with a beneficial response in four.
Two recent reviews of sarcoidosis mentioned reports of patients being treated with immunosuppressive medications including methotrexate, azathioprine, chlorambucil, and (investigationally) cyclosporin A. The total number of patients was less than 100 and, except for limited use in patients with dermal sarcoidosis, these medications are reserved for use in patients who have critical organ involvement and who have failed a course treatment with steroids. Benefit was seen in the majority of patients. Chronic sarcoidosis has been likened to an immunologic disorder and, superficially, immune-altering drugs would be expected to be efficacious. However, Kinney et al could find no precise mechanism(s) for the beneficial results.
The lack of response to corticosteroid therapy in my patient with bilateral pleural and pericardial effusions prompted both the pericardiectomy and the use of cyclophosphamide. The lack of any systemic markers for her disease has proved distressing since her disease activity can only be assessed by the appearance of pleural effusion or by the changes on the thallium and gallium scans. These objective indices, however, have shown marked improvement and there has been a complete subjective response. Equally distressing is the inability to reduce the drug dosage to less than 50 mg per day. Thus, cyclophosphamide may prove beneficial to the patient with sarcoidosis. Caution must be used in patient selection since there exists concern that cyclophosphamide, even in low doses, may have neoplastic potential.