High Prevalence of Myocarditis Mimicking Arrhythmogenic Right Ventricular Cardiomyopathy

Differential Diagnosis by Electroanatomic Mapping-Guided Endomyocardial Biopsy

Patient population and noninvasive studies

We studied 30 consecutive patients (17 male, mean age 43 ± 17 years) admitted to our institution from January to December 2006, with a diagnosis of ARVC on the basis of noninvasive evaluation, including 12-lead electrocardiogram, 24-h Holter monitoring, signal-averaged electrocardiogram, 2-dimensional echocardiography, and contrast-enhanced cardiac magnetic resonance. The latter was not performed in 4 patients because of claustrophobia.

The disease was considered familial in the presence of other family members with autopsy or biopsy-proven disease or premature (<40 years) sudden death at pedigree analysis.

After noninvasive study, all patients fulfilled current criteria for diagnosis of ARVC defined by the European Society of Cardiology and International Society and Federation of Cardiology Task Force (7).

Invasive study

The invasive study was approved by the Institutional Review Board, and all patients gave their written informed consent. All patients underwent coronary and left and right ventricular angiography (right and left anterior oblique views), 3-dimensional EAM, EMB, and programmed ventricular stimulation.

In particular, RV angiography, representing the gold standard for the detection of wall motion abnormalities and aneurysms, was performed before EAM to provide the RV silhouette in 2 views, thus improving the anatomical accuracy of EAM. On the basis of EAM, EMBs were drawn from areas presenting electrical abnormalities, as detailed in the following text.

EMB

Right ventricular EMB (4 to 5 samples from each patient) were obtained through the femoral vein with the use of a pre-formed long sheath and a disposable Bioptome (Cordis, Johnson & Johnson, Miami, Florida). Once EAM was completed, the mapping catheter's tip was directed against abnormal voltage areas, and the distal end of the sheath was placed close to it. Sheath position was checked in right and left anterior oblique views (Fig. 1),and then biopsies were withdrawn from wall segments with abnormal voltage. In case of normal EAM, EMB were withdrawn from conventional sites including apex and interventricular septum.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

EAM-Guided Endomyocardial Biopsy

(A)The electroanatomic mapping (EAM) catheter (arrowheads)is maintained in contact with a right ventricle free wall segment presenting low voltages. The distal end of the pre-formed sheath (arrows)is positioned close to the sheath's tip (right anterior oblique view 30°). (B)The correct position of the pre-formed sheath close to the mapping catheter is confirmed in other views (left anterior oblique view 45°). (C)Endomyocardial biopsies are drawn close to the mapping catheter's tip from the low voltage area. Arrowindicates the opened Bioptome against the ventricular wall (right anterior oblique view 30°).

In 2 patients with abnormal voltage areas mainly involving the basal posterior wall, the Bioptome was introduced through a steerable introducer (Agilis NxT, St. Jude Medical, Minnetonka, Minnesota) to reach that segment.

Histology and immunohistochemistry

Two to 3 samples were processed for histology and immunohistochemistry. For histology, multiple 5-μm-thick sections were cut and stained with hematoxylin-eosin, Miller's elastic Van Gieson, and Masson's trichrome, and examined by light microscopy. Immunohistochemistry for the characterization of inflammatory infiltrates was performed using the following antibodies: CD3, CD8, CD45RO, CD68 (Dako Corporation, Glostrup, Denmark), as previously described (10–12). To quantify the inflammatory infiltrates, CD8+ and CD45RO+ positive lymphocytes were counted per high-power field (400-fold magnification) in all available fields, and the mean number was calculated, as previously described (10–12).

In patients presenting histological evidence of fibrofatty replacement, a histomorphometric analysis was performed on Masson's trichrome-stained sections to calculate the extent of myocardial atrophy and fibrofatty replacement. Images obtained at ×5 magnification with a digital camera (Leica DFC 420C, Leica Microsystems, Wetzlar, Germany) were stored as TIFF files and analyzed with a dedicated imaging software (Leica Application Suite version 3.0, Leica Microsystems) to calculate the percent area occupied by adipose tissue, replacement fibrosis, and residual myocardium.

The diagnosis of myocarditis was based on Dallas criteria and immunohistochemistry (15); in particular, a T-lymphocyte infiltration (>7/mm²) in the presence of cytotoxic (CD8+) and activated (CD45RO+) lymphocytes was considered diagnostic (16). The diagnosis of ARVC was made on the basis of extensive fibrofatty myocardial atrophy with a percentage of fat >3% and fibrous tissue >40% associated with amounts of residual myocytes <45% of the specimen at morphometric analysis (17).

Molecular biology studies

Two frozen myocardial specimens per patient were used to detect the presence of cardiotropic viruses by polymerase chain reaction, as previously described (10,12).

Statistical analysis

Continuous variables are expressed as mean ± SD. Categorical differences between groups were evaluated by the 2-tailed Fisher exact text. Differences between group means were compared by the unpaired ttest. All probability values reported are 2 sided, and a probability value <0.05 was considered statistically significant. Survival free of life-threatening ventricular arrhythmias was analyzed by the Kaplan-Meier method, and comparison between groups performed by the log-rank test. Statistical analysis was performed with the SPSS version 13.0 software package (SPSS Inc., Chicago, Illinois).

Clinical features

Clinical characteristics and instrumental findings of the patients are summarized in Tables 1 and 2.⇓Before tissue characterization was obtained, all patients fulfilled current diagnostic criteria, and all but 1 presented at least 1 major criterion. A family history was documented in 4 patients, and clinical symptoms were present in 20, including 1 subject resuscitated from cardiac arrest. The time interval from onset of symptoms to the invasive study ranged from 1 to 36 months (mean 19 ± 31 months).

Ventricular arrhythmias with left bundle branch block morphology were documented in all patients, including sustained monomorphic ventricular tachycardia in 9, nonsustained ventricular tachycardia in 12, and frequent premature ventricular beats in 9. Arrhythmias were symptomatic in 12 patients with ARVC and in 8 patients with myocarditis; in asymptomatic patients, ventricular arrhythmias were documented at Holter monitoring performed after detection of frequent ventricular ectopic beats at resting electrocardiogram. At the time of the invasive study, antiarrhythmic therapy included sotalol in 4 patients and beta-blockers in 15 patients.

All patients presented structural and functional RV abnormalities at echocardiography, cardiac magnetic resonance imaging, and/or angiography. In particular, wall motion abnormalities were present in 26 subjects, with microaneurysms in 7 subjects. Cardiac magnetic resonance showed areas of RV delayed enhancement in 13 of 26 patients, evidence of RV fatty replacement in 10 of 26 patients, and LV delayed enhancement in 8 of 26 patients (Table 2).

EAM and programmed ventricular stimulation

The mean number of sites sampled in RV EAM was 155 ± 36. Twenty-nine (97%) of 30 patients had an abnormal voltage map. All patients had >1 area (mean 2.4 ± 1.0) with contiguous bipolar electrograms with voltage values <0.5 mV (scar tissue) surrounded by a larger zone with signal amplitudes between 0.5 and 1.5 mV indicating abnormal myocardium.

Right ventricular outflow tract and anterior free wall were the most frequently involved segments, whereas interventricular septum presented abnormal voltages in 1 patient only. Abnormal voltage areas were focal in 12 patients with main involvement of the outflow tract and inferior and posterior wall, and the remaining patients had widespread RV involvement usually associated with dilation and contractile impairment. Interestingly, wall motion abnormalities corresponded in all cases to abnormal voltage areas.

Programmed ventricular stimulation induced sustained monomorphic ventricular tachycardia in 9 patients and ventricular fibrillation in 3 patients.

EMB

In 15 patients the presence of myocardial atrophy and fibrofatty replacement confirmed the diagnosis of ARVC. In the remaining 15 cases, histology showed the presence of inflammatory infiltrates associated with necrosis of adjacent myocytes, consistent with the diagnosis of active myocarditis according to Dallas criteria. In all cases, immunohistochemistry showed inflammatory infiltrates to be mainly represented by activated and cytotoxic T lymphocytes. No patients showed histological features of sarcoidosis, granulomatous, and/or giant cell myocarditis. In patients with myocarditis, none of the samples examined presented areas of fatty replacement; moreover, no inflammatory infiltrates were observed in association with fibrofatty replacement in ARVC patients. Molecular biology studies demonstrated the presence of viral genome in 5 patients with myocarditis (parvovirus B19 in 3 cases, influenza virus in 2 cases) and in none with ARVC (p = 0.04).

Safety of EAM-guided endomyocardial biopsy

In all patients, EMB was performed by the same operator (M.P.). No major complications, including cardiac tamponade or perforation, were observed. Echocardiography performed immediately and 24 h after the procedure failed to show pericardial effusion in all cases, including 1 patient having chest pain after EMB. No complications at the site of femoral access, no sustained ventricular arrhythmias, atrial fibrillation, or development of atrioventricular or complete right bundle branch block during the execution of EAM and EMB were documented.

Comparison between ARVC and myocarditis group

The comparison between patients with biopsy-proven ARVC and patients with histological evidence of myocarditis failed to detect any difference in terms of clinical presentation, electrocardiography, arrhythmic profile, and imaging studies. A higher prevalence of family history of sudden death, although statistically not significant, was observed in the ARVC group as expected. Structural and functional abnormalities of the RV, detected either at angiography or cardiac magnetic resonance, were similar in the 2 groups, including the presence of RV delayed enhancement, whereas evidence of RV fatty infiltration was more frequent in ARVC patients (Tables 1 and 2). Similarly, the presence of abnormal voltage areas at 3-dimensional EAM as well as the induction of arrhythmias during programmed ventricular stimulation were not able to distinguish the 2 myocardial disorders (Figs. 2 and 3).⇓⇓The patients showing a normal map and the patients showing involvement of the interventricular septum were both affected by focal myocarditis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Noninvasive and Invasive Findings in a Representative Patient With Histological Diagnosis of ARVC

(A)Twelve-lead electrocardiogram shows negative T waves from V₁to V₃and frequent premature ventricular beats with left bundle branch block/inferior axis morphology. (B)Twelve-lead electrocardiogram shows a ventricular fibrillation induced during programmed ventricular stimulation. Right anterior oblique view (30°) of right ventricle (RV) angiography (C)and bipolar voltage map (D)show a moderately enlarged and hypokinetic RV with increased trabeculation of diaphragmatic and free wall and ectasia of the infundibular segment. Electroanatomic map shows an extensive low-voltage area (red= voltage <0.5 mV) of the free wall extending from the inferobasal segment to the outflow tract and infundibular segments. The arrowindicates the area where endomyocardial biopsies (EMBs) were drawn from. (E)Right ventricular EMB shows extensive myocardial atrophy with fibrofatty replacement consistent with the diagnosis of arrhythmogenic right ventricular cardiomyopathy (ARVC). (Masson's trichrome, original magnification ×5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Noninvasive and Invasive Findings in a Representative Patient With Histological Diagnosis of Myocarditis

(A)Twelve-lead electrocardiogram showing negative T waves from V₁to V₄and frequent premature ventricular beats, even in triplets, with left bundle branch block/inferior axis morphology. (B)Twelve-lead electrocardiogram showing sustained ventricular tachycardia with left bundle branch block morphology. Right anterior oblique view (30°) of right ventricle (RV) angiography (C)and bipolar voltage map (D)showing a moderately enlarged and hypokinetic RV with ectasia of the free wall and outflow tract and an extensive low-voltage area (red= voltage <0.5 mV) of the free wall extending from posterolateral to anterolateral and outflow tract segments. The arrowindicates the area where endomyocardial biopsies (EMBs) were drawn from. (E)Right ventricular EMB shows diffuse inflammatory infiltrates associated with necrosis of adjacent myocytes consistent with the diagnosis of active myocarditis. (Hematoxylin and eosin, original magnification ×100).

Treatment and follow-up

On the basis of clinical features and pathological findings, 13 patients with biopsy-proven ARVC received an implantable cardioverter-defibrillator (ICD). Two patients refused ICD implantation and were treated with antiarrhythmic drugs including amiodarone and beta-blockers. Among patients with myocarditis, 1 patient presenting with syncope and inducible ventricular tachycardia received an ICD, and the remaining patients were all treated by beta-blockers.

At a mean follow-up of 21 ± 8 months (range 18 to 26 months), 7 (47%) patients with ARVC experienced recurrence of symptomatic sustained ventricular arrhythmias, in all cases interrupted by appropriate ICD interventions (direct current shocks in 4 cases, pacing in 3). All patients with myocarditis remained asymptomatic and free from arrhythmic events, and no shocks were documented in the patient with the ICD (Fig. 4).Two-dimensional echocardiography failed to show progression of the disease in terms of right and left ventricular dimensions and function in both groups.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Kaplan-Meier Analysis of Arrhythmic Event-Free Survival Depending on the Histological Findings

Arrhythmic events were defined as symptomatic sustained ventricular arrhythmias and/or appropriate implantable defibrillator shocks. Solid line= arrhythmogenic right ventricular cardiomyopathy; dashed line= myocarditis. Log-rank p = 0.003.

Comparison with previous studies

Endomyocardial biopsy and autopsy findings have clearly demonstrated that myocarditis is a frequent cause of ventricular arrhythmias in persons with apparently normal heart or minimal electrocardiographic and cardiac structural abnormalities (12,18–20). Myocarditis may selectively affect the RV causing structural abnormalities, including microaneurysms (9–12), and arrhythmic manifestations typical of ARVC (9,11,21). In fact, myocardial inflammatory infiltrates associated with myocyte necrosis and replacement fibrosis may lead to functional and structural changes of RV myocardium resembling those produced by fibrofatty replacement, and representing the substrate of the abnormal voltage map and ventricular arrhythmias. In this regard, it has been recently reported that RV sarcoidosis may mimic EAM and arrhythmic features of ARVC (22).

At variance with a previous study (12), in our series EAM failed to distinguish ARVC from myocarditis, as we observed abnormal voltage areas in patients with myocarditis as well. This discrepancy is probably due to the shorter interval from the onset of arrhythmias to EMB in our study, reducing the possibility of nonspecific histological findings associated with a spontaneous resolution of the inflammatory process. In addition, EAM guidance, allowing us to obtain myocardial samples from affected areas, reduced the sampling error of EMB. In fact, myocarditis, as well as ARVC, is frequently a focal disorder and can be missed at biopsy, particularly when a limited number of EMBs is drawn from the same conventional site. Noteworthy in a previous study, EMBs obtained by a traditional approach were nondiagnostic in 25% of patients with abnormal EAM (9). These data suggest a significant increase of EMB sensitivity under EAM guidance.

Moreover, the evidence of normal EAM in patients with histological evidence of focal inflammatory cardiomyopathy reported by Corrado et al. (9) supports the notion that, in patients with myocarditis, voltage abnormalities as well as ventricular arrhythmias are mostly related to the persistence of myocardial inflammation rather than to scar tissue.

Clinical and therapeutic implications

The differential diagnosis between RV myocarditis and ARVC has relevant clinical implications regarding the management of arrhythmias and the prevention of sudden death. Current treatment strategies suggest the implantation of an ICD in most symptomatic patients with ARVC and even in asymptomatic patients presenting risk factors such as family history and severe right ventricular involvement or positive programmed ventricular stimulation. Conversely, no guidelines concerning the treatment of life-threatening ventricular arrhythmias in the context of active myocarditis are currently available. Several studies suggest that, in these patients, ventricular arrhythmias are associated with the persistence of the inflammatory process, are generally self-limited, and exhibit a benign course (9–12) usually not requiring major therapeutic strategies including defibrillator implantation. Accordingly, in the present study, an ICD was implanted in 13 of 15 patients with a histological diagnosis of ARVC, but only 1 patient with myocarditis received a defibrillator because of syncope and inducible ventricular tachycardia. Recurrent symptomatic sustained ventricular arrhythmias with appropriate defibrillator therapy were documented in 7 patients with ARVC. In contrast, the outcome of patients with myocarditis was excellent, thus confirming the benign prognosis of ventricular arrhythmias in this subset of patients. These findings, although obtained in a small selected population with midterm follow-up, suggest that the identification of the underlying myocardial disease may represent adjunctive useful information to define the therapeutic approach to young persons with life-threatening ventricular arrhythmias and may warrant the execution of a complete invasive study.

It is needless to emphasize that inappropriate ICD implantation, beside costs, may have a devastating effect on life-style, in particular for young patients, due to the risk of inappropriate shocks and infection related to multiple device substitution. Of note, in a recent study, patients with the Brugada syndrome showed a rate of inappropriate shocks far exceeding the appropriate ICD interventions (23), with youngest age representing a risk factor for inappropriate activation due to sinus tachycardia and over-sensing of muscular activity.

Study limitations

Although dealing with a rare condition, the relatively small sample size of study population and the short duration of the follow-up may represent limitations of the present study. In addition, as our institution is a tertiary center for the study of cardiomyopathies, many patients with clinical suspicion of ARVC were sent from other hospitals, and a possible referral bias may have influenced the features of study population and the time of enrollment. Genetic studies for known desmosomal gene mutations were not performed as they are not necessary for the diagnosis of ARVC. Further studies comparing electroanatomical, histological, and genetic features in larger populations could clarify other aspects of the pathogenesis of ARVC. With regard to the safety of the invasive procedure, we must acknowledge that the high experience level of the cardiologist performing the biopsies could have minimized the risks related to this new approach, which should, therefore, be considered in referral centers with experience on endomyocardial biopsy.