Tubercular Myocarditis Presenting With Ventricular Tachycardia
Rohit Khurana; Joseph Shalhoub; Anju Verma; Ravi Assomull; Sanjay K. Prasad; Jaspal S. Kooner; Amarjit Sethi Nat Clin Pract Cardiovasc Med. 2008;5(3):169-174. ©2008 Nature Publishing Group
Summary and the Case
Background: A previously fit and healthy 30-year-old man
reported experiencing palpitations accompanied by nausea, sweating and
presyncope. These symptoms were found to be associated with episodes of
nonsustained ventricular tachycardia. He was a nonsmoker, did not drink
excessively, denied illicit drug use and had no family history of
structural cardiac disease or sudden death.
A 30-year-old Indian-born man, currently resident in the UK, presented to the emergency department complaining of sudden-onset palpitations that had persisted throughout the previous night and disturbed his sleep. This episode was associated with nausea, sweating and presyncope, but no chest pain or respiratory symptomatology. The palpitations had first occurred 4 months earlier and had been increasing in frequency since then. His general practitioner had prescribed atenolol (25 mg once daily) 3 weeks previously, but there had been no improvement in symptoms, and the patient reported experiencing a further one or two episodes per week until presentation. He was a nonsmoker, drank up to 28 units of alcohol per week and denied illicit drug use. There was no family history of structural cardiac disease or sudden death. Before these episodes he had been fit and well, and had no history of anorexia or weight loss.
There were no abnormal findings on clinical examination and S3 and S4 heart sounds were absent. The patient's blood pressure was 132/80 mmHg. Electrocardiography demonstrated ventricular ectopy but was otherwise unremarkable. The results of hematological and biochemical tests were normal; notably thyroid function and troponin levels were within normal limits. He was admitted to the coronary care unit, where electrocardiographic monitoring recorded episodes of nonsustained ventricular tachycardia (VT) that correlated with the palpitations (Figure 1). Amiodarone therapy was commenced (200 mg 3 times a day).
An electrophysiological study performed 7 days after admission induced a slow VT (90 beats/min) with an inferior axis right bundle branch block morphology which would terminate and reinduce spontaneously. Ablation was not performed. Multiple excitatory maneuvers to induce the clinical VT were unsuccessful; this failure was attributed to the ongoing treatment with amiodarone and atenolol. These drugs were discontinued with a view to repeating the electrophysiological investigation 2 months later. Soon after discontinuation of therapy, however, the patient experienced frequent and multiple episodes of nonsustained VT as well as a number of prolonged runs—two of which necessitated direct current cardioversion. Amiodarone therapy was recommenced along with bisoprolol (10 mg once daily). Despite this medication, the patient had a further episode of prolonged VT (without hemodynamic compromise) that did not respond to lidocaine (1.5 mg/kg bolus followed by 2 mg/min infusion) and also required direct current cardioversion. Mexelitine (200 mg twice daily) was added to the patient's therapeutic regime.
Tumor markers were within normal limits and HIV test results were negative. Serum angiotensin-converting enzyme levels were mildly raised (69 units/l; normal range 8-52 units/l), but 24 h urinary calcium excretion was normal (0.87 mmol/24 h; normal range 0.5-7.5 mmol/24 h). The patient's erythrocyte sedimentation rate was 30 mm/h.
Physical examination revealed that there were no palpable cervical, supra-clavicular, axillary or inguinal lymph nodes available for excision or biopsy. Imaging investigations demonstrated significant right-sided mediastinal lymphadenopathy ( Box 1 ; Figures 2 and 3). Mediastinoscopy and lymph-node biopsy were performed, and microscopy demonstrated a large number of confluent granulomas with Langhans-type multinuclear giant cells and minimal areas of necrosis. Ziehl Nielsen staining identified a low number of acid and alcohol fast bacilli. A scant growth of mycobacterium tuberculosis was cultured, and shown to be fully sensitive to conventional antituberculous drugs. A sputum culture was negative for acid and alcohol fast bacilli. A Heaf test was also performed but proved inconclusive.
A standard antituberculous regimen was commenced, comprising rifampicin (120 mg), isoniazid (50 mg), pyrazinamide (300 mg), ethambutol (1 g) and pyridoxine (10 mg; all taken once daily). A steroid regime of methylprednisolone was also prescribed (1 g administered intravenously once daily for 3 days, followed by 1 g taken orally once daily). Cardiac MRI, repeated 2 weeks later, continued to show significant fibrosis (Figure 2). The short T1 inversion recovery signals were, however, less active than on the previous scan, suggesting that myocardial active inflammation was reduced. An automatic implantable cardioverter-defibrillator (ICD) was fitted, which precluded further MRI.
The patient's palpitations improved significantly over the following 6 months of antituberculous therapy and his erythrocyte sedimentation rate normalized to 6 mm/h. At the most recent 18-month follow-up he had completed his course of therapy and been weaned off steroids. Interrogation of the ICD confirmed that it had not activated once since implantation, and there had been no need for antitachycardia pacing. Repeat CT confirmed the complete resolution of mediastinal lymphadenopathy (Figure 3). Treatment with amiodarone and mexiletine were discontinued, but bisoprolol was maintained.
Discussion of Diagnosis
Cardiac tuberculosis is a rare but well-recognized condition that was first reported many decades ago. It is estimated that 1% of all cases of tuberculosis have cardiac involvement. Before the introduction of chemotherapy the overall incidence of cardiac tuberculosis, as detected by autopsy, was less than 0.3%. This condition most commonly affects the pericardium, while endocardial, myocardial, valvular or coronary arterial involvement is exceedingly rare.
The case presented in this report is a typical example of myocardial involvement in tuberculosis. An anatomical predilection for the right-sided mediastinal lymph nodes has been described in this condition, making the right side of the heart the most vulnerable area of the myocardium owing to the potential for direct spread. Although not involved in the present case, the right atrium is particularly susceptible to disease spread. Tubercular myocarditis usually results from direct hematogenous seeding from the overlying pericardium or from lymphatic spread via the mediastinal lymph nodes, as in this case. It is unclear why the spread of tuberculosis displays a lower affinity for the myocardium than the pericardium. Three different types of tubercular myocarditis have been reported: diffuse infiltrative, military, and nodular with central caseation—known as a tuberculoma. The MRI images with short T1 inversion recovery strongly suggest that the patient presented in this report had the diffuse infiltrative variant. The patient also had echocardiographic evidence of mild concentric left ventricular hypertrophy, but this was attributed to longstanding mild and untreated hypertension and considered an incidental finding that was unrelated to the primary diagnosis.
Tubercular myocarditis can remain clinically asymptomatic (with a diagnosis being made at autopsy), or can present with sudden cardiac death,[4,5,6] intractable ventricular arrhythmias, long QT syndrome, heart block or congestive heart failure. Sudden cardiac death related to tuberculosis is rare, and is usually attributable to bronchopneumonia and complications of pulmonary cavitation leading to massive haemoptysis. Cardiac complications, such as ventricular tachyarrhythmias, are rarer still, but are potentially treatable and prompt recognition of the underlying cause is crucial in these cases.
The increasing rates of immigration from countries where tuberculosis and multidrug-resistant variants are endemic emphasize the importance of making a prompt diagnosis in cases of tubercular myocarditis. Once the condition is suspected, relevant investigations must be undertaken to exclude viral illnesses, sarcoidosis and malignancies such as lymphoma. Definitive diagnosis of the condition relies on the detection of characteristic histological changes using Ziehl Nielsen staining. Endomyocardial biopsy was considered at an early stage in the present case but it is not a procedure without risk and was, therefore, delayed until the results of the mediastinoscopic lymph-node biopsy were available. This investigation proved sufficient to confirm the diagnosis, despite the presence of only minimal numbers of acid and alcohol fast bacilli. Had lymph-node biopsy been inconclusive, consent for an endomyocardial biopsy would have been requested. For postmortem analysis, identification of the Mycobacterium tuberculosis DNA complex using a polymerase chain reaction is the diagnostic gold standard.
The value of cardiac MRI and late gadolinium enhancement for investigating infiltrative myocardial pathologies is evolving. Isolated case reports in the literature describe the use of MRI in the characterization of cardiac tuberculomas, but the role of imaging in these conditions remains largely ill-defined.[13,14] Serial MRI would have been informative in the present case and would have been conducted had an ICD not been implanted. Follow-up CT imaging was, therefore, performed as an alternative. The value of serial MRI was conveyed by a second case of tubercular myocarditis involving a 56-year-old man who was admitted with an acute coronary syndrome associated with VT (R Khurana, unpublished data). Images taken at presentation demonstrated late enhancement and increased signal in the anteroseptal wall of the basal left ventricle, but a diagnosis of lymph-node tuberculosis was not established until MRI was performed 5 months later—prompting a definitive biopsy to be performed (Figures 4 and 5). MRI was repeatedly carried out as follow-up, and successfully demonstrated progressive diminution and eventual resolution of the lesion that had been noted on the initial scan. These cases emphasize the difficulties inherent in making a diagnosis of tubercular myocarditis, and it is possible that the condition could be underdiagnosed antemortem.
Treatment and Management
The literature on antemortem cases of myocardial tuberculosis is limited and treatment largely follows empirical guidance. The index of suspicion for making this diagnosis should be raised if the arrhythmias prove refractory to the additive effects of multiple antiarrhythmic drugs. The choice of which β-blocker to prescribe should ideally be made on the basis of the drug's efficacy for the treatment of ventricular arrhythmias, but their relative merits have not been validated in this clinical setting. Antituberculous therapy should be initiated once the diagnosis is confirmed by histology but, as suggested by the present case, these tests might not always be definitive. The ideal duration of therapy is also currently unclear, although one previous report documented successful resolution of VT following a year of antituberculous and antiarrhythmic therapy.
There is little published evidence on the efficacy of adjunctive corticosteroid therapy in the treatment of myocardial tuberculosis, making its prescription—as issued in the present case—controversial. It is well recognized, however, that myocardial sarcoidois, a potentially treatable condition that can masquerade as tuberculosis, is responsive to steroid treatment. Cardiac transplantation has been described as a successful treatment in a case of isolated asymmetrical septal hypertrophic cardiomyopathy secondary to myocardial tuberculosis, which presented as severe ventricular dysfunction.
The patient presented in this report was fitted with an ICD in order to protect against the recurrence of further arrhythmic episodes. The duration of this therapy is under active review, and eventual removal of the device will depend upon the continued absence of symptoms and ventricular arrhythmias, as assessed by routine interrogation.
Tubercular myocarditis is a rare condition that should be suspected in patients with arrhythmias who could have been exposed to tuberculosis. It is important that awareness of this condition remain high so that a prompt diagnosis can be made in future cases. Early endomyocardial biopsy is indicated, where possible, if clinical suspicion is strong and adjunctive imaging is suggestive. Antituberculous therapy can result in successful regression of the disease and termination of malignant arrhythmias.
Box 1. List of Diagnostic Imaging Studies Performed on the Patient and Summary of the Relevant Findings
Rohit Khurana, Department of Cardiology, Imperial College Healthcare NHS Trust, St Mary's Hospital, Praed Street, London W2 1NY, UK. Email: email@example.com
R. Khurana is a Specialist Registrar, J. Shalhoub and A. Verma are Senior House Officers, J.S. Kooner is a Professor and Consultant and A. Sethi is a Consultant at the Department of Cardiology, Ealing Hospital NHS Trust, Southall, and R. Assomull is a Research Fellow and S.K. Prasad is a Consultant at the Cardiovascular Magnetic Resonance Unit, Royal Brompton Hospital, London, UK.
Competing Interests: The authors declared no competing interests.