The clinical diagnosis of arrhythmogenic right ventricular cardiomyopathy (ARVC) is often challenging due to phenotypic variation, reduced/age-related penetrance, and lack of a diagnostic test. A single report has suggested quantitative myocardial immunoanalysis for the desmosomal protein plakoglobin as a diagnostic test with high sensitivity and specificity. We performed immunohistochemistry for plakoglobin and a control protein on myocardial biopsies with fibrofatty replacements from 50 consecutive, unrelated patients. The clinical, genetic, and immunohistochemical data were evaluated by independent observers in a blinded manner. The immunohistochemical and clinical diagnoses were compared and the sensitivity, specificity, and predictive values calculated. Our analysis showed 37 samples (74%) with a reduced immunosignal for plakoglobin. Of the 34 patients with a clinical diagnosis of ARVC, 29 displayed a reduced plakoglobin signal. Of the 14 patients with a clinical diagnosis other than ARVC, 6 displayed a reduced signal. Two patients were excluded from further analysis. A sensitivity of 85%, a specificity of 57%, a positive predictive value of 83%, and a negative predictive value of 62% were found. In conclusion, immunohistochemical analysis for plakoglobin, applied as a diagnostic test for ARVC, seems associated with a relatively high sensitivity, but limited specificity, and although additional validation is required, we advocate caution in basing clinical decision-making on the proposed diagnostic test.
Quantitative immunohistochemical myocardial analysis for plakoglobin has been suggested as a new diagnostic test for arrhythmogenic right ventricular cardiomyopathy (ARVC). In a study from 2009, the investigation of myocardial samples from patients with various clinical diagnoses showed that a low immunoreactive signal for plakoglobin was a sensitive and specific sign of ARVC. The finding was observed in biopsy specimens from all areas of the heart and regardless of whether fibrofatty replacements were present. The method was validated with both an immunofluorescence and an immunoperoxidase setup.
As only a single report has investigated the role of immunostaining for plakoglobin as a diagnostic test for ARVC, we found it of interest to apply the method to the patients referred to our center. Our aim was to perform consecutive evaluation of the plakoglobin immunoreactive levels of all available myocardial biopsies with fibrofatty replacement and correlate the immunohistochemical findings with the clinical and genetic information using a double-blinded approach.
Methods
All endomyocardial biopsies received at the Department of Pathology, Rigshospitalet, Copenhagen University Hospital, Denmark from 2000 to 2009 were evaluated (n = 3,980). The institution is a third-line referral center. All biopsies diagnosed with fibrofatty replacement in the ventricular myocardium (n = 50) were included in the present study. Fibrofatty replacement was defined as histological demonstration of fibrosis and adipose tissue. The extent of myocardial atrophy was not estimated, and morphometric analysis was not used. The biopsies were obtained as a part of routine clinical management for suspected heart disease. The intended biopsy location was the right side of the interventricular septum and ≥3 biopsies were obtained per catheterization. Either jugular or femoral access was used.
As control material, biopsies from donor hearts (n = 10) with no clinical or pathologic evidence of heart disease were used.
For all patients, clinical data were obtained from existing records. Age, gender, and personal and family history were noted. The results of available diagnostic testing, including electrocardiography, 2-dimensional echocardiography, Holter monitoring, stress electrocardiography, signal-averaged electrocardiography, magnetic resonance imaging, right ventriculography, and endomyocardial biopsies were registered. From these data, the patients were classified as fulfilling or not fulfilling the 1994 Task Force criteria. If not, the clinical diagnosis was noted. The evaluation of clinical data and classification were performed by experienced cardiologists blinded to the immunohistochemical data.
A subgroup of the patients had been screened for mutations in known ARVC-associated genes ( DSC2 , DSG2 , DSP , JUP , PKP2 , TGFb3 , TMEM43 ), and available genotypes were also registered.
All biopsies consisted of formalin-fixed, paraffin-embedded cardiac tissue. The sections were cut at a thickness of 3 μm, mounted on glass slides, deparaffinized, rehydrated, heated in TRS buffer (High pH/K8002, Dako, Glostrup, Denmark) for 20 minutes at 97°C, and endogenous peroxidase activity was blocked. The samples were incubated with monoclonal mouse antiplakoglobin antibody (Sigma Aldrich, Dorset, United Kingdom) followed by incubation with Envision (Dako) peroxidase-labeled polymer with Real DAB+ (Dako) as a chromogen substrate. To determine the optimal dilution of the antibody, we performed initial immunostaining on normal myocardium using a stepwise dilution technique with dilutions from 1:4,000 to 1:300,000. No discernible difference in the distribution of plakoglobin at the intercalated disks was observed until the antibody was diluted 1:300,000. With this dilution, plakoglobin was strongly reduced or absent in all control biopsies. Accordingly, anti-plakoglobin antibody diluted 1:50,000 and 1:250,000 was applied on all endomyocardial biopsies included in the present study. Quantification of the immunoreactive intensity was performed in areas composed of structurally normal myocardium. A reduced plakoglobin signal level was defined as >50% reduction in either the distribution or the intensity of the immunostained areas compared to the control samples. All samples were evaluated by 2 independent experienced pathologists. Both were blinded to clinical and genetic data.
To exclude time-dependent tissue decay, we performed control staining for dystrophin using the monoclonal mouse antibody NCL-DYSA (diluted 1:100, NovoCastra, Newcastle upon Tyne, United Kingdom).
Based on the blinded clinical and immunohistochemical classification, the sensitivity, specificity, and negative and positive predictive values of immunohistochemical staining for plakoglobin, as a diagnostic test for ARVC, were calculated.
Results
Endomyocardial biopsies from of 50 unrelated patients (24 men) were included in the study. The mean age at the time of biopsy was 42 years (range 21 to 63). A clinical diagnosis of ARVC according to the published criteria was made in 34 patients. Two patients presented with cardiac arrest due to ventricular fibrillation and had fibrofatty infiltrations in the myocardium. Both patients died before the diagnostic workup was completed and were therefore classified as having “probable ARVC.” The remaining 14 patients had the following diagnoses: right ventricular outflow tract tachycardia (RVOT-VT; 6 patients), dilated cardiomyopathy (2 patients), heart transplant recipients (2 patients), Takotsubo cardiomyopathy (1 patient), polymyositis (1 patient), idiopathic ventricular fibrillation (1 patient), and premature ventricular controls without identifiable underlying structural heart disease (1 patient). The clinical and immunohistochemical data are summarized in Table 1 .
| Clinical Diagnosis | Patients (n) | Plakoglobin Reduced |
|---|---|---|
| Arrhythmogenic right ventricular cardiomyopathy | 34 (68%) | 29 (85%) |
| Probable arrhythmogenic right ventricular cardiomyopathy | 2 (4%) | 2 (100%) |
| Right ventricular outflow tract tachycardia | 6 (12%) | 2 (33%) |
| Dilated cardiomyopathy | 2 (4%) | 2 (100%) |
| Heart transplant recipients | 2 (4%) | 1 (50%) |
| Takotsubo cardiomyopathy | 1 (2%) | 1 (100%) |
| Polymyositis | 1 (2%) | 0 (0%) |
| Idiopathic ventricular fibrillation | 1 (2%) | 0 (0%) |
| Premature ventricular contractions | 1 (2%) | 0 (0%) |
Twenty patients from the cohort had been genotyped. In 14 (70%) of the 20 patients, no potential pathogenic mutation was identified. In 6 patients, a total of 9 mutations were identified (3 patients were digenic carriers) consisting of 3 mutations in DSC2 , 3 mutations in PKP2 , 1 mutation in DSG2 , 1 mutation in JUP , and 1 mutation in TMEM43 .
In 37 of 50 endomyocardial biopsies, a marked reduction in the immunoreactive levels of plakoglobin was seen at the intercalated disks. Thirteen biopsies had plakoglobin levels inseparable from those of the controls. Of the 34 patients with a clinical diagnosis of ARVC, 29 (85%) had diminished plakoglobin levels. No unique phenotypic manifestations were seen in the 5 patients with a normal plakoglobin signal. The 2 patients with probable ARVC both showed reduced plakoglobin levels. Of the 6 patients with a clinical diagnosis of RVOT-VT, the plakoglobin levels were reduced in 2 (33%). Of the remaining 8 patients, a reduction in the plakoglobin level was seen in the biopsies from 4 patients (50%). Immunostaining for the control protein dystrophin showed a comparable signal in all samples. Representative images from trichrome-stained myocardial biopsies, plakoglobin immunoreactions, and controls are shown in Figure 1 .
