Active Lymphocytic Myocarditis Treated with Murine OKT3 Monoclonal Antibody

Abstract

This report describes the case of a 33-year-old woman with biopsy-proven, active lymphocytic myocarditis manifested by intractable ventricular tachycardia, nonspecific intraventricular block, and myocardial dysfunction. We treated her successfully with OKT3 monoclonal antibody and antiarrhythmic agents. Immunosuppression is not recommended in patients with infectious or postinfectious myocarditis. However, it may have an important role in autoimmune myocarditis. In the few reports in the medical literature that we were able to find, OKT3 monoclonal antibody was administered early, in the setting of acute, fulminant autoimmune myocarditis. Our patient received OKT3 therapy in a later phase of the disease, when inflammatory infiltrates were accompanied by extensive fibrosis and severe damage of cardiomyocytes.

Our patient had concomitant Helicobacter pylori infection and a strong positive family history of gastric cancer, a disease often associated with H. pylori. We discuss the possibility of a causal relationship between H. pylori infection and autoimmune myocarditis. (Tex Heart Inst J 2002;29:113–7)

Active myocarditis presenting as a malignant arrhythmia with deteriorating myocardial function is a serious, frequently fatal condition. Approaches to management include symptomatic treatment, ¹ immunosuppression, ^2–5 placement of a ventricular assist device for hemodynamically compromised patients, ^3,6,7 or, not infrequently, heart transplantation. ⁸

A growing number of reports show links between Helicobacter pylori infection and autoimmune disease. ^9,10 We report the occurrence of biopsy-proven, active lymphocytic myocarditis in a patient with concomitant H. pylori infection and a strong positive family history of gastric cancer, whose autoimmune myocarditis we treated with OKT3 monoclonal antibody.

Case Report

In October 1999, a 33-year-old woman was admitted to her local hospital because of palpitations. On evaluation, her blood pressure was 110/80 mmHg, her heart rate was 150 beats/min, and she had no symptoms of congestive heart failure. Electrocardiography revealed wide QRS tachycardia at 150 beats/min (left bundle branch block pattern, left-axis deviation). After intravenous administration of Ajmalin^* and metoprolol, sinus rhythm was restored, with nonspecific intraventricular block. Her chest radiograph revealed mild cardiomegaly and normal pulmonary markings; her echocardiogram showed mild left ventricular enlargement and mild hypokinesis of the anterior wall of the left ventricle.

Because the patient was the mother of 3 children and was breast-feeding her youngest child (an 18-month-old girl), she was reluctant to leave her hometown for examination by specialists in Warsaw. There was no personal or family history of heart disease or autoimmune disease, nor was there a personal history of an influenza-like syndrome or rash in the weeks preceding the onset of symptoms. However, the patient had a 10-year history of peptic ulcer disease and a strong family history of gastric cancer: her sister had undergone surgery for gastric cancer at the age of 23 years, and her father had died of gastric cancer at the age of 46.

During the next 4 weeks of local hospitalization, the patient had several episodes of ventricular tachycardia, which lasted from 2 to 12 hours despite optimal dosage of amiodarone and treatment with metoprolol. These episodes of arrhythmia were well tolerated hemodynamically. The patient also had dyspepsia, and gastroscopy revealed H. pylori-associated gastric inflammation. Eradication was done with triple therapy (omeprazole, metronidazole, and amoxicillin) for 1 week.

Following this 4-week period of recurrent ventricular tachycardia, the patient agreed to be transferred to our referral center for thorough evaluation. On the day of admission, she was asymptomatic, her blood pressure was 100/60 mmHg, and she had a regular pulse of 70 beats/min. Auscultation revealed a ventricular gallop and clear lungs. There was no evidence of hepatomegaly or edema. Electrocardiography revealed sinus rhythm at 70 beats/min and nonspecific intraventricular block, with an elongated, corrected QT interval (446 ms), possibly due to amiodarone treatment. Chest radiography revealed mild cardiomegaly and normal pulmonary markings. On echocardiography, there was evidence of mild left ventricular dilatation (left ventricular end-diastolic dimension, 59 mm) and globally depressed left ventricular ejection fraction (30%), right ventricular dilatation, mild mitral and tricuspid insufficiency, and no pericardial effusion. Upon 24-hour Holter electrocardiographic monitoring, she was found to have single ventricular ectopy (242/24 h) and no complex arrhythmia. With a low CRP level (<7 mg/L), a sedimentation rate of 8 mm/hr, and normal blood cell counts, the patient had no markers of systemic inflammatory process. Tests for antinuclear antibodies and rheumatoid factor were negative.

In the apparent absence of acute inflammatory disease and arrhythmia at rest, this patient was judged not to be in the acute phase of the disease, which by then had a 5-week history of symptoms. Accordingly, we decided to perform an exercise stress test on a treadmill before proceeding with coronary angiography, endomyocardial biopsy, and, eventually, electrophysiology. At 4 METS (metabolic equivalents), the patient developed a run of ventricular tachycardia (120 beats/min), which was followed by a ventricular rhythm of 110 beats/min with frequent capture beats; this was well-tolerated hemodynamically. She reverted to sinus rhythm after intravenous administration of 2 amiodarone boluses (75 mg each).

Heart catheterization was performed on the next day. Catheter manipulation induced ventricular arrhythmia; measurements done at a ventricular rhythm of 110 beats/min revealed mild pulmonary artery hypertension, moderate pulmonary venous hypertension, and mild systemic venous hypertension. The angiographic left ventricular ejection fraction was 20%, and coronary angiography was normal; a tissue sample of the right ventricular endomyocardium was taken for histologic and immunohistochemical biopsy. After heart catheterization, the patient was transferred to our coronary care unit (CCU) for cardioversion. After 2 intravenous boluses of amiodarone (150 mg each) were administered, cardioversion (200 watt seconds) was performed twice. This restored sinus rhythm for a while, before reversion to ventricular tachycardia. The tachycardia at rates up to 150 beats/min was characterized by left-axis deviation, and the tachycardia at 130 beats/min was marked by right-axis deviation (both with frequent capture beats); occasionally, ventricular rhythm was observed at 110 beats/min, with frequent fusion beats (Fig. 1). Despite continuous intravenous infusion of xylocaine, metoprolol, boluses of magnesium sulfate, and therapeutic levels of amiodarone, the patient's ventricular tachycardia was intractable.

Fig. 1 Types of sustained ventricular arrhythmia recorded in the patient: A) ventricular tachycardia at 150 beats/min, with left-axis deviation and left bundle branch block pattern, shown with a capture beat; B) ventricular tachycardia at 130 beats/min, with right-axis deviation, shown with a capture beat; and C) ventricular rhythm at 110 beats/min with fusion beats.

The endomyocardial biopsy revealed intense inflammatory infiltrate, along with prominent fibrosis and destruction of cardiac myocytes (Figs. 2A and 2B); immunohistochemical analysis identified CD4+ lymphocytes (Fig. 2C) and a few macrophages. Furthermore, expression of class II human lymphocyte antigens was found both on endothelial cells and on cardiac myocytes (Fig. 2D), which suggested autoimmune myocarditis. Detailed examination of all histologic slides for the presence of giant cells found none.

Fig. 2 Right ventricular endomyocardial biopsy: A and B) Histologic sections show destruction of cardiac myocytes, inflammatory cell infiltrates, and prominent fibrosis. H&E, orig. ×200 (A); ×160 (B); C) immunohistochemical examination demonstrates CD4+ lymphocytes in the infiltrate. Orig. ×160; D) immunohistochemical examination shows expression of class II human lymphocyte antigens on endothelial cells, inflammatory cells, and some cardiac myocytes. Orig. ×160.

After seeing the endomyocardial biopsy results, we decided to induce immunosuppression immediately, with oral cyclosporine and intravenous boluses of methylprednisone. The high doses of methylprednisone caused deterioration of the patient's hemodynamic status; accordingly, we administered dobutamine for inotropic support and consulted a cardiac surgeon in regard to the potential need of a ventricular assist device. The cyclosporine caused vomiting and had to be discontinued 2 days later.

Because of the patient's incessant ventricular tachycardia, we considered therapy with the murine OKT3 monoclonal antibody. On the 5th day of continuous ventricular tachycardia, the patient gave written informed consent, and we began a course of intravenous Orthoclone OKT3 (Ortho Biotech Inc.; Raritan, NJ), 5 mg daily for 5 days. The patient was premedicated with ranitidine (100 mg IV) and acetaminophen (500 mg orally) before each dose; clemastine (2 mg IV) was administered once daily. For the first 2 doses, the patient received 200 mg of hydrocortisone intravenously. Fever, dyspnea, and chills were detected with the first 2 doses. On the 6th day of incessant ventricular tachycardia at the rate of 140–160 beats/min, we administered a 250-mg bolus of bretylium tosylate intravenously, followed by continuous infusion at 2 mg/min. On the 8th day of the arrhythmia, the ventricular tachycardia slowed to 120 beats/min, and cardioversion reverted the patient to sinus rhythm; we immediately discontinued the bretylium tosylate. Then the patient's blood pressure increased to 130/80 mmHg, and it was possible to discontinue dobutamine support; however, her left ventricular ejection fraction decreased from 30% on admission to 15% to 20%.

After discontinuation of OKT3 therapy, maintenance immunosuppression was used. On the 3rd day after restoration of sinus rhythm, the patient left the CCU. Thereafter, consequent to depletion of circulating T-lymphocytes by OKT3 monoclonal antibody, she developed opportunistic infections. Oral and vaginal candidiasis responded favorably to treatment with nystatin. Ten days after the onset of OKT3 administration, the patient developed fever and a nonproductive cough, followed by nausea, diarrhea, and elevated liver transaminase levels, which resolved with symptomatic treatment. Because the symptoms were highly suggestive of cytomegalovirus (CMV) infection, retrospective serial evaluation of CMV immunoglobulin M and G titers was done and revealed that the patient had been positive for CMV IgG-antibodies before administration of the OKT3 monoclonal antibody; after OKT3 therapy, CMV immunoglobulin M seroconversion was transient, followed by a 10-fold rise in CMV immunoglobulin G titers. During the 5th week of the hospital treatment, the patient's condition began to improve, and her left ventricular ejection fraction rose to 25%. A 2nd endomyocardial biopsy was not performed, due to the risk that catheter manipulation would induce arrhythmia in our very sick patient. She was discharged on the 45th hospital day after a symptom-limited exercise test, which did not induce ventricular tachycardia. Serial 24-hour electrocardiographic tracings were negative. Her medication at the time of discharge was carvedilol (25 mg daily), amiodarone (200 mg daily), and cilazapril (1 mg daily). The causative agent of the disease remains unclear; serial search for antibodies against common viruses yielded negative results. At her 5-month follow-up, the patient was active as a housewife, taking care of her 3 children, while in NYHA functional class II with a left ventricular ejection fraction of 30% and no ventricular tachycardia on 24-hour Holter monitoring.

Discussion

This case shows that patients with unexplained, sustained ventricular tachycardia associated with myocardial dysfunction should be promptly evaluated with an endomyocardial biopsy. Administration of OKT3 monoclonal antibody, along with the antiarrhythmic agents, was helpful in our patient. Clinical and echocardiographic improvement at the 5-month follow-up argues against the continued presence of an active inflammatory process in the myocardium. In the few reports in the medical literature, OKT3 monoclonal antibody was administered early, in patients who had fulminant myocarditis. ^2,3 Our patient received OKT3 therapy in the later phase of the disease, when inflammatory infiltrates were accompanied by extensive fibrosis and severe damage of cardiomyocytes.

Although the administration of OKT3 monoclonal antibody leads to direct depletion of T lymphocytes, few serious complications of the treatment have been reported, chief among them opportunistic infections. ¹ We observed, as have others, ² candida infection and CMV disease without sequelae. The risk of symptomatic CMV infection suggests the need for standard evaluation of CMV serologic status before administration of OKT3 monoclonal antibody; CMV sero-positive patients could then be targeted for antiviral prophylaxis. ¹¹ Other complications of OKT3 therapy, seen in heart transplant recipients, are lymphoproliferative disorders. ¹²

The insidious course of disease in our patient with recurrent ventricular tachycardia was not typical of a severe inflammatory process in the myocardium. In the literature, there have been several reports of a fulminant course of myocarditis with severe inflammation. ^3,4,6 Slower onset, refractory ventricular tachycardia, and conduction-system disease leading to progressive deterioration might suggest giant-cell myocarditis; ^13–15 however, in our patient no giant cells were found on detailed examination of biopsy specimens. Immunosuppression is not recommended in patients with infectious or postinfectious myocarditis; however, it may have an important role in autoimmune myocarditis. ¹⁶ The presence of T-helper lymphocytes in the infiltrate and the expression of class II human lymphocyte antigens on some cardiac myocytes in our patient suggested an autoimmune process. ^17,18

Another interesting issue is whether H. pylori infection in our patient, who had a strong family history of disease potentially associated with H. pylori, had any causal relation to the inflammatory process in the myocardium. Helicobacter pylori is a common pathogen, affecting more than 70% of the population world-wide. Molecular mimicry may be involved in the pathogenesis of autoimmune myocarditis: cross-reactive epitopes between cardiac myosin and infectious agents have been identified. ¹⁹ Accordingly, it would be interesting to learn from databases of large cardiology centers whether patients with autoimmune myocarditis, especially that manifesting with recurrent ventricular tachycardia and conduction system disease, had H. pylori infection.

In summary, recurrent, sustained ventricular tachycardia associated with myocardial dysfunction may be an important feature of active lymphocytic myocarditis. Definitive diagnosis is possible only with endomyocardial biopsy. In patients who present early with this condition, empirical immunosuppressive therapy may be considered. Treatment with OKT3 monoclonal antibody remains an unproven but attractive option. Whether or not H. pylori infection in our patient was a contributing factor remains an open question.

Addendum

Beginning in August 2001, this patient was treated with 2 antiarrhythmic agents to prevent recurrence of her ventricular tachycardia: amiodarone (200 mg) and prajmalin (40–60 mg) daily. Prajmalin,^* a derivative of ajmalin, is used for chronic, oral administration, in a maximal dose of 20 mg every 6 to 8 hours. During the follow-up period, this patient's left ventricular ejection fraction was stable at 25%–30%.

In the beginning of March 2002, catheter ablation of the recurrent ventricular tachycardia enabled discontinuation of the prajmalin. On 9 April 2002, the patient died of stroke in her local hospital, on the 1st postoperative day after an elective cholecystectomy. We cannot exclude the possibility that cholestasis in our patient was caused by prolonged treatment with prajmalin. No autopsy was performed.