A rare, but frequently fatal, side effect of the antipsychotic drug clozapine is myocarditis. We report a case of hypersensitivity myocarditis secondary to clozapine administration that was diagnosed in vivo for the first time through endomyocardial biopsy and was successfully treated with corticosteroids. Histologic diagnosis was based on the evidence of eosinophilic infiltration of the endomyocardium and eosinophil degranulation. Endomyocardial biopsy was performed in order to establish or exclude a clear-cut relationship between cardiac dysfunction and clozapine, and was crucial to establish a correct diagnosis and appropriate treatment. Clozapine withdrawal and targeted 8-day, low-dose corticosteroid therapy resolved the symptoms and restored cardiac function.
Key words: differential diagnosis; drug hypersensitivity; endomyocardial biopsy; myocarditis; schizophrenia
Abbreviations: LV = left ventricle, ventricular
Clozapine is the "gold standard" treatment for schizophrenia patients who are resistant to neuroleptics. Its use, limited by the well-known agranulocytosis risk, has also been associated with severe cardiovascular side effects and sudden death. (1-3) Both dilated cardiomyopathy and myocarditis, as result of direct toxicity and drug hypersensitivity, respectively, have been described at autopsy. (2,3) We report a case of hypersensitivity myocarditis secondary to clozapine administration diagnosed for the first time in vivo by endomyocardial biopsy and successfully treated with corticosteroids.
A 27-year-old chronic schizophrenic man (according to criteria of the Diagnostic and Statistical Manual of Mental Disorders IV), who was resistant and intolerant to neuroleptic drugs was admitted for the first time to our psychiatric ward because of severe psychotic symptoms, including a delusion of gender transformation and self-defeat behavior. After stopping treatment with haloperidol, risperidone, and biperiden, which were ineffective and burdened the patient with severe akathisia, clozapine therapy was started. The dose was increased up to 250 mg/d for a period of 11 days, while concomitant therapy with lorazepam was tapered from 6 to 3 mg/d. A clozapine plasma concentration of 425 ng/mL was found on day 12, when a full antipsychotic response was appreciated. On the same day, a persistent fever (temperature, 38.5[degrees]C) associated with pharyngodynia and neutrophilic leukocytosis appeared. Within 72 h, the clinical picture worsened, despite antibiotic treatment for a suspected infectious disease. The patient showed severe malaise and dyspnea, while the chest radiogram showed a slight enlargement of cardiac silhouette. Clozapine-related myocarditis or acute viral myocarditis following a throat infection then was suspected, clozapine therapy was withdrawn, and the patient was transferred into the ICU.
On ICU admission, physical examination revealed tachycardia (heart rate, 130 beats/min) with gallop rhythm, and BP was 100/60 mm Hg. An ECG showed sinus tachycardia with diffuse ST-segment and T-wave abnormalities. The erythrocyte sedimentation rate was 47 mm/h (normal range, 2 to 10 mm/h), the C-reactive protein level was 205.4 mg/L (normal range, 2.0 to 6.0 mg/L), and the blood cell count showed leukocytosis (14 x [10.sup.9]/L) with 7.0% eosinophils. The troponin I level was slightly increased (1.65 ng/mL; normal level, < 1.5 ng/mL). A two-dimensional echocardiogram showed a mild left ventricular (LV) dilatation (LV end-diastolic diameter, 57 mm) with a markedly reduced contractility (ejection fraction, 35%). Serology findings for the most common cardiotropic viruses, including hepatitis C virus and HIV, and the results of immunologic studies were negative.
Four days later, after the patient gave informed consent, cardiac catheterization with ventriculography, coronary angiography, and an LV endomyocardial biopsy were performed in order to establish a clear-cut relationship between cardiac dysfunction and clozapine. This information was essential in deciding whether to resume the patient's most successful antipsychotic treatment. The results of coronary angiography were normal, and the LV was diffusely hypokinetic. Six good-sized LV endomyocardial samples were drawn and processed for histology and polymerase chain reaction for the most common cardiotropic viruses. (4) Histology in all samples showed the presence of extensive inflammatory infiltrates, mainly represented by degranulated eosinophils and lymphocytes, which often were adherent to injured myocytes (Fig 1), infiltrating the endocardium with the focal apposition of eosinophil-rich thrombotic material. Polymerase chain reaction results for cardiotropic viruses was negative.
[FIGURE 1 OMITTED]
On the basis of the histologic findings and molecular biology study results, myocarditis was treated with prednisone for a short period (1 mg/kg/d for 4 days tapered to 0.33 mg/kg/d for the following 4 days) in order to limit a possible worsening of psychosis. After 8 days of treatment, the patient had a visible improvement of symptoms, with a reduction of heart rate to 75 beats/min and the disappearance of a gallop rhythm. An ECG showed an increase of voltages with the disappearance of repolarization abnormalities. The echocardiogram showed a reduction of LV dimensions (LV end-diastolic diameter, 52 mm), with marked improvement of LV contractility (ejection fraction, 65%). The patient was discharged from the ICU without therapy with cardiac drugs, while the persisting psychotic remission allowed a prolongation of antipsychotic washout before starting a new treatment. Cardiac function improvement was maintained after corticosteroids withdrawal and after three months follow-up.
We have reported the first endomyocardial biopsy detection of clozapine-induced hypersensitivity myocarditis. The histologic diagnosis was based on the evidence of eosinophilic infiltration of the endomyocardium, and on eosinophil degranulation with the likely release of the cationic protein that ultimately mediates both myocyte damage and endocardial thrombus formation.
Along with clozapine, several other drugs have been associated with hypersensitivity myocarditis, including sulfonamides, penicillin, tetracycline, streptomycin, diuretics, methyldopa, and amitriptyline. Unfortunately, drug hypersensitivity with a reaction confined to the heart is difficult to recognize, so that the majority of the reports are indeed postmortem observations. (2)
In our patient, clozapine therapy was clearly indicated, and had a rapid and impressive efficacy, therefore drug withdrawal required clear evidence of an iatrogenic cause of myocarditis. Time course, treatment schedule, and clinical presentation corresponded to the common presentation of this idiosyncratic reaction. (2,3) Nevertheless, the clinical manifestation of fever associated with cardiac dysfunction and increase in troponin I level are nonspecific, and in our case could well have been due to a viral infection or a toxic mechanism. Corticosteroids would have been ineffective in treating a case of toxic damage, and deleterious in the presence of viral myocarditis, as they can promote viral replication and spreading in myocardial tissue. (4) Conversely, corticosteroid efficacy in patients with hypersensitivity and autoimmune myocarditis has been established, (5) with expected benefits clearly overcoming the known risk of psychotic relapse. Although we cannot completely exclude the idea that clozapine withdrawal alone would have provided cardiac recovery, the additional curative role of steroidal therapy is supported by the persistence of severe cardiac dysfunction in the week following drug withdrawal and preceding corticosteroid treatment.
In our case, endomyocardial biopsy was crucial to establish a correct diagnosis and appropriate treatment, allowing a prompt recovery of the patient's cardiac compromise. Early recognition, instrumental diagnosis, and related specific treatment made this patient's course favorable.
Clozapine-induced hypersensitivity myocarditis should be suspected by psychiatrists and physicians when cardiac dysfunction appears suddenly, and appropriate diagnostic and therapeutic strategies must be undertaken promptly.
(1) La Grenade L, Graham D, Trontell A. Myocarditis and cardiomyopathy associated with clozapine use in the United States. N Engl J Med 2001; 345:224-225
(2) Killian JG, Kerr K, Lawrence C et al. Myocarditis and cardiomyopathy associated with clozapine. Lancet 1999; 354: 1841-1845
(3) Hagg S, Spigset O, Bate A et al. Myocarditis related to clozapine treatment. J Clin Psychopharmacol 2001; 21:382-388
(4) Frustaci A, Chimenti C, Calabrese F et al. Immunosuppressive therapy for active lymphocytic myocarditis: virological and immunologic profile of responders versus nonresponders. Circulation 2003; 107:857-863
(5) Frustaci A, Chimenti C, Pieroni Met al. Giant cell myocarditis responding to immunosuppressive therapy. Chest 2000; 117:905-907
* From the Cardio-Thoracic and Vascular Department (Drs. Pieroni and Chimenti) and the Department of Neuropsychiatric Sciences (Drs. Cavallaro and Smeraldi), San Raffaele Hospital, "Vita-Salute" University School of Medicine, Milan, Italy; and the Cardiology Department (Dr. Frustaci) Catholic University, Rome, Italy.
Manuscript received January 28, 2004; revision accepted June 15, 2004.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: email@example.com).
Correspondence to: Andrea Frustaci, MD, FCCP, Cardiology Department, Catholic University, Largo A. Gemelli 8, 00168 Rome, Italy; e-mail: firstname.lastname@example.org
COPYRIGHT 2004 American College of Chest Physicians
COPYRIGHT 2004 Gale Group