Arrhythmogenic right ventricular cardiomyopathy: severe structural alterations are associated with inflammation
- Oscar Campuzano1,2,
- Mireia Alcalde1,
- Anna Iglesias1,2,
- Catherine Barahona-Dussault2,
- Georgia Sarquella-Brugada3,
- Begoña Benito2,
- Dabit Arzamendi2,
- Jose Flores2,
- Tack Ki Leung2,
- Mario Talajic2,
- Antonio Oliva4,
- Ramon Brugada1,2
- 1Centre de Genètica Cardiovascular Institut Investigació Biomèdica Girona and University of Girona, Girona, Spain
- 2Montreal Heart Institute, University of Montreal, Montreal, Canada
- 3Secció Cardiologia, Hospital Sant Joan de Déu, Barcelona, Spain
- 4Institute of Forensic Medicine, Catholic University, Rome, Italy
- Correspondence to Dr Ramon Brugada Terradellas, Director Cardiovascular Genetics Center IDIBGI-UdG, Dean School of Medicine, Universitat de Girona, Girona 17003, Spain;
- Accepted 4 August 2012
- Published Online First 3 September 2012
Aim Arrhythmogenic Right Ventricular Cardiomyopathy (ARVC) is a rare cardiomyopathy associated with sudden cardiac death. It is characterised by a progressive right ventricle (RV) fibrofatty replacement, although biventricular replacement (BV) is also common. Inflammation believed to be a key player in disease progression and outcome. Our study investigates the relationship between the presence of inflammatory infiltrates in myocardium and the severity of structural heart alterations in ARVC.
Methods Our study included eight control and 36 ARVC postmortem human heart samples. We performed macroscopic assessment and microscopic analysis for different inflammatory cell types.
Results Fibrofatty replacement of RV was present in all our cases. Thirteen cases showed sole RV involvement (36.11%). Of these, only one showed inflammatory infiltrates (7.69%). Sixteen cases showed severe ARVC phenotypic forms characterised by BV involvement and right auricular (RA) fatty accumulation plus RV dilation (44.44%); eight of them also showed inflammatory infiltrates (50%). Immunohistochemical studies revealed ventricular multifocal inflammatory infiltrates, showing seven T-lymphocytes as the main infiltrate cell types.
Conclusions The presence of inflammatory infiltrates in ventricular myocardium of ARVC samples is associated with severe structural heart changes, indicating that an inflammatory process may be a modulator of severity in ARVC.
Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare cardiomyocyte disease,1 characterised by adipose and fibrous tissue replacement of the right ventricle (RV) in a so-called triangle of dysplasia.2 Biventricular involvement (BV) can be found in up to 50% of ARVC cases, characterised by RV and left ventricle (LV) involvement, mainly the apical and inferior areas.3
The prevalence has been estimated to be from 1/1000 to 1/5000 in the general population, but this could be underestimated due to non-diagnosed or misdiagnosed cases.3 ARVC causes myocardial electrical impulse impairment inducing clinical symptoms which may range from palpitations to syncope.4 Disease expression is highly variable and incomplete, weakening both the diagnostic process and clinical management, mainly at early disease stages.5 ,6 In addition, ARVC is a leading cause of sudden cardiac death (SCD) in people under 35 years of age, and accounts for up to 10% of deaths from undiagnosed cardiac disease in the <65 age group.7 It has also been associated with 3%–4% of SCD in young male athletes, considering exercise as a common precipitant of arrhythmias.8
ARVC has been reported as a familial disease although with a wide range of incidence (15%–70%).7 Most of the cases follow an autosomic-dominant Mendelian pattern, although some cases have been reported as autosomic recessive.9 Multiple mutations have been described in nine different genes, mainly encoding desmosomal proteins.10 However, ARVC may be explained by three theories, all compatible with the genetic origin11: (1) Apoptosis in RV may follow the disruption of the intercalated discs of cardiomyocytes due to mutations in genes encoding proteins of the desmosome,12 (2) Inflammation: certain gene defects may induce immune alterations contributing to myocarditis due to a viral infection, causing myocyte apoptosis, replacement and subsequent inflammation.13 ,14 (3) Transdifferentiation: cardiac progenitor cells in myocardium can change into fibrous or fatty cells.15 A study collaboration of our group demonstrates that adipocyte replacement is originated from the second heart field cardiac progenitors.16
Thus, a combination of genetic and environmental factors appears to play a role in the pathogenesis of the disease. Despite this, it is unclear what the role of inflammation and apoptosis is, although both inflammatory infiltrates and apoptotic cells are common findings in ARVC.17 Because of multivarious mechanisms that mediate both findings and their possible causal relationship, our focus was to clarify the inflammatory role in ARVC macroscopic heart defects. Hence, mast cells, macrophages, neutrophils and T-lymphocytes are the most common inflammatory cells recruited to inflammatory diseases,18 such as ARVC.19 However, it is not known whether the inflammatory cell infiltration is a cause of cell death, or the consequence of infective/immune mechanisms.20 ,21 Likewise, it is not known whether the inflammatory process is associated with a severe tissue form of the disease, playing a pathogenic role in arrhythmogenesis. In this regard, the aim of our study was to characterise postmortem ARVC samples to perform a clinicopathologic correlation. Thus in this study we aim to determine 1) the presence of inflammatory inflitrates in the cardiac tissue, 2) the inflammatory cell type and, 3) the relationship between the presence of inflammatory infiltrates and the severity of structural myocardial alterations, defined as BV involvement, right auricular (RA) external fatty accumulation with RV dilation and even left auricular (LA) external fatty accumulation.
Our retrospective study includes 36 heart samples diagnosed with certain ARVC after postmortem examination, and disclosed no other causes of death. In addition, no history of coronary artery disease, diabetes or hypertension was reported in any of them. All samples were collected in our institution and evaluated according to postmortem measures of the International Society and Federation of Cardiology Task Force criterion.8 ,22 The control group included eight heart samples of patients who died due to non-cardiac causes, and confirmed negative ARVC diagnosis after autopsy. The purpose of including a control group was to exclude macroscopic and microscopic common alterations described in healthy populations. The Montreal Heart Institute (Montreal, Canada) institutional review board, and the ethics committee, approved all procedures performed in our study.
Macroscopic study included measurement of whole heart weight and ventricular wall thickness. The anatomical assessment was performed by an expert forensic pathologist, and confirmed by another independent pathologist. Heart regions were methodically examined to identify the possible presence of regional/diffuse fibrofatty accumulation, mainly in ventricle/auricle walls. In addition, extent and distribution of ventricular myocardium atrophy were confirmed by routine wall transillumination technique. Representative full-thickness myocardium sections (anterior, posterior and lateral) of both ventricles and auricles were fixed in formalin and included in paraffin blocks. For each section (paraffin block), parallel series 10 microns thick were obtained with a microtome and mounted on slides. In both ventricles and auricles, random slices of each section were labelled with Masson Trichrome stain. Additionally, Sudan Black B and Picro-sirius Red staining were also carried out to assess the fatty and fibrous tissue, respectively.
To evaluate the presence of inflammatory infiltrates, five non-consecutive slides of each section were labelled in both ventricles for the same immunohistochemical marker (see below) to identify lymphocyte T, lymphocyte B and neutrophils. Staining non-consecutive slides avoid the possible double labelling of one cell in two or three parallel slides. Additionally, immunolabelling for mast cells and macrophages was also carried out. Paraffin sections were rinsed thoroughly in tris buffered saline (TBS) (0.05 M Trizma base, 150 mM NaCl and pH 7.4). Following rehydration, the sections were immersed in 10 mM sodium citrate (antigen retrieval, S1699 Dako) and boiled at 95°C for 30 min, followed by cooling at room temperature for 20 min. After that, endogenous peroxidase was blocked with 2% H2O2 in 70% methanol. After rinsing, sections were treated with blocking buffer (BB) (10% fetal calf serum in TBS+1% triton X-100–TBS+T–) for 30 min, and incubated overnight at 4°C, and then for 1 h at room temperature with primary antibody diluted in BB (anti-CD3 for T-lymphocyte, anti-CD20 for B-lymphocyte, anti-myeloperoxidase -MPO- for neutrophils, antimacrophages and antitryptase for mast cells) (table 1). Sections were rinsed and incubated for 1 h at room temperature with the corresponding secondary antibody diluted in BB (table 1). After incubation, sections were rinsed with TBS+T and incubated in avidin-peroxidase (P-0364, Dako) (1 : 400 in BB) for 1 h at room temperature. Finally, the peroxidase reaction product was visualised in 100 ml of TB (0.05 M Trizma base, pH 7.4) with 50 mg of 3′-diaminobenzidine (DAB) and 35 μl of hydrogen peroxide. Negative controls were incubated without primary antibodies (data not shown). After washing, the slices were mounted on gelatin-coated slides, air-dried, dehydrated in alcohol, cleared in xylene, and cover slipped with DPX. Parallel slices from all controls and ARVC samples were studied through a Nikon Eclipse 50i microscope interfaced to a DS-2Mv camera and a HP4600 PC. Pictures of representative areas were taken at different magnifications using the NIS-Elements BR3.0 software (Nikon). Both in control and ARVC samples, five slices for each labelling and section (anterior, posterior and lateral) were analysed in RV and LV. Only profiles clearly identified as specifically labelled cells were evaluated. Researchers were blinded to sample identity when microscopic analysis was performed. Anatomical landmarks were used to ensure that parameters were analysed at similar levels within and between samples. Statistical analysis includes χ2 distribution (χ2-test) to determine significance between independent variables. A value of p<0.05 was considered statistically significant.
Control samples did not show any macroscopic structural alteration suspicious of ARVC after postmortem analysis. Only a very low diffuse fatty presence in atrial and ventricular exterior walls was identified, as it is well described in normal hearts.23 ,24 All hearts had a normal weight (300–350 g in males and 250–300 g in females), and thickness of right ventricular free wall (left 13–15 mm and right 3–5 mm). Microscopic studies did not reveal replacement of myocardium, neither right/left ventricles, nor atrial. Inflammatory immunohistochemical studies showed monocytes/macrophages, most of them inside blood vessels. A reduced presence of mast cells, neutrophils and T-lymphocytes was also observed only inside blood vessels. In myocardial tissue, only scattered macrophages were identified (data not shown). All these results in our control group help us to determine the common structural changes in healthy hearts in comparison with changes related to the disease.
Our study includes 36 ARVC samples (21 men and 15 women; age range from 10 to 52 years; mean age, 28) (figure 1). All of them were Caucasian individuals. No clinical presentation of ARVC and/or myocarditis was registered in any of our patients before death.
Anatomic/macroscopic analysis revealed weight of the hearts to be between 220 and 617 g (mean 383.4 g), in concordance with previous ARVC studies.14 ,25 In addition, and also in concordance with ARVC parameters, all hearts showed less than 3 mm thickness in RV free wall (mean 2.4 mm). All samples showed high accumulation of fatty tissue in RV, and 23 (63.8%) also showed fatty tissue in left ventricles (BV involvement), in contrast with normal low fat accumulation present on the epicardial surface,23 ,24 or simple fatty infiltration in the RV associated with chronic alcohol abuse, inherited myopathies or normal ageing process in healthy subjects.26 In addition, and also in concordance with ARVC parameters, samples with BV involvement showed less than 10 mm thickness in LV free wall (mean 8.9 mm). Microscopic analysis confirmed RV fibrofatty replacement in all our ARVC cases, and also in LV of all 23 cases with BV fibrofatty replacement (table 2). In addition, myocardial contraction band necrosis was identified in fibrofatty replacement areas, revealing myocardial cell death, a hallmark in ARVC. There was a coexistence of normal and partially degenerated myocardial fibres that potentially could provide the substrate for slow conduction and re-entrant ventricular arrhythmias. Anatomical evaluation also identified that 16 of these 23 samples showed external BV fatty accumulation plus right atrial (11 samples), or external BV plus biatrial fatty accumulation (five samples) (table 2) (figure 2A,B). Microscopic analysis in atrial sections did not identify microscopic fibrofatty accumulation. Therefore, our study identified presence of inflammatory infiltrates in the most severe forms of the disease, characterised by BV involvement, RA and even LA fatty accumulation. A total of 22 samples showed no inflammatory infiltrates (table 2), all of them with RV replacement, but only 10 of 22 showed BV involvement. We also identified that two samples without inflammatory infiltrates showed BV involvement, RA and LA fatty external accumulation.
Inflammatory assessment revealed RV macrophages inside blood vessels in all ARVC samples, similar to our control samples. Inside the myocardium, 14 samples showed foci of inflammatory infiltrates, mainly in areas with extensive fibrofatty replacement. We identified a mix of macrophages, neutrophils and mast cells in RV of all of them. In addition, seven samples also showed T-lymphocytes (figure 3). Curiously, only one case of isolated RV disease showed inflammatory infiltrates (7%) (macrophages, neutrophils, mast cells and T-lymphocytes). More interestingly, 42.8% of BV samples displayed inflammatory infiltrates, and the percentage rose to 62.5% when the fibrofatty involvement went beyond the ventricles and into the atria (table 2). Thus, the inflammatory cells were more present as there was more cavity involvement (tables 3 and 4). After comprehensive analysis, there was no clear correlation between the severity of the disease and the cell types identified. There were seven cases of main presence of T cells (neutrophils, and less mast cells and macrophages are also observed), and seven samples with mainly neutrophils, and less mast cells and macrophages (without T cells). All samples showed neither B-lymphocytes nor plasma cells (table 2).
Concerning the ARVC samples without inflammation, our study revealed similar labelling of inflammatory cells than in the control group. So, in both these groups we identified mainly monocytes/macrophages, most of them inside blood vessels. In some samples, also a reduced existence of mast cells, neutrophils and T-lymphocytes were identified, but only inside blood vessels. Thus, inside myocardial tissue, similar labelling of scattered macrophages was identified in both groups.
ARVC is an extremely complex disease with variable evolution and outcome which induces a wide diversity of phenotype severities.3 Most of published ARVC populations have been diagnosed after genetic screening and/or after clinical symptoms during lifetime. Our group has been diagnosed after autopsy, with no previous ARVC clinical history known. Gender prevalence and mean age at death, even at young age,27 were in concordance with previous ARVC studies.28 It has been also reported that normal fat replacement on the epicardial surface of RV,23 ,24 and simple fatty infiltration in the RV are associated with chronic alcohol abuse, inherited myopathies, or normal ageing process in healthy subjects.26 All our samples differ from these situations, showing ARVC characteristics confirming disease affectation. Therefore, our results can be compared with other ARVC cohorts because our samples have been diagnosed following current international autopsy protocols.
ARVC is considered a progressive disease which affects RV, but BV involvement is also common29 and frequently associated with worst clinical arrhythmias.7 Numerous ARVC samples showed fibrofatty BV replacement, all of them associated with severe anatomical defects, also suggesting a worst clinical phenotype. In addition, recent data correlates younger (<40 years) SCD victims with RV plus extensive LV involvement.30 In agreement, we identify both a young range of ages, (only one sample greater than 50 years), and most of our BV involvement samples are less than 40 years.
The presence of inflammatory infiltrates surrounding foci of necrotic or degenerative myocytes has been reported in ARVC patients,3 as also occurring in our samples. We identified myocardial contraction band necrosis and inflammatory infiltrates in fibrofatty replacement areas. In addition, in ARVC patients, the presence of inflammatory infiltrates31 and necrosis32 appears to be the leading mechanisms associated with increased risk of arrhythmias and a worst disease outcome, although the underlying mechanism remains unknown. Though, there has not yet been a correlation between the presence of inflammatory infiltrates and the anatomical severity of the disease. We addressed this key question, and our study identified presence of inflammatory infiltrates in the most severe forms of the disease, characterised by BV involvement, RA and even LA fatty external accumulation. The inflammatory cell infiltration was identified mainly in RV.30 These data may help in the diagnostic evaluation of ARVC patients when RV endomyocardial biopsy is performed,33 despite RV biopsy limitations as a diagnostic tool because of the patchy nature of the disease.34 Despite all these data showing that inflammatory infiltrates are correlated with severe forms of the disease, two samples without inflammatory infiltrates showed BV involvement, RA and LA fatty external accumulation. This fact is in concordance with published ARVC series, supporting that other factors (such as viral infection) also can modify phenotype outcome.
It is well known that myocarditis is mainly induced by a viral infection which induces T-lymphocytes presence in myocardium.35 ,36 Presence of inflammatory cell infiltrates in biopsies without fibrofatty replacement has been described, consistent with an inflammatory cardiomyopathy mimicking ARVC.37 It has been published that ARVC cases showed adipose tissue in contrast with reported myocarditis cases.38 We identified T-lymphocytes in seven of our samples, all them with fibrofatty ventricular replacement, confirming ARVC diagnosis. Thus, there still exists disagreement about whether ARVC may result from a primary myocardial alteration as the consequence of an inflammatory necrotic injury followed by a fibrofatty repair.25 However, the inflammation could be a secondary event not implicated in ARVC pathogenesis, because an already diseased myocardium can have an increased susceptibility to viral infection21 ,39 with resulting myocardial inflammation that can eventually play a role in ARVC clinical symptoms and increasing risk of arrhythmias.31 ,40 Finally, the inflammatory presence in RV may be also due to desmosomal dysfunction.3 Thus, it remains to be elucidated whether desmosomal dysfunction induces myocyte cell death and subsequent ventricle inflammation or, in contrast, whether inflammation due to viral infection induces myocyte apoptosis, which is followed by desmosomal dysfunction.
There are limitations that need to be acknowledged. First is the small number of patients. Major cohorts of ARVC patients should be analysed in order to corroborate our results. Second is the limitation of DNA purification from our samples. Many protocols for DNA extraction from paraffin-embedded tissue (PET) have been tested, including the best PET-DNA extraction method which gives an average of only two-thirds of the region of interest,21 but the DNA obtained was exceedingly degraded. This fact implies, as previously indicated in the discussion, an important restriction concerning the lack of genetic studies to eventually detect viral genome. It was recently suggested that ARVC and myocarditis requires analysis of biopsy specimens in order to be distinguished.41 Third is a limitation related to DNA purification is the absence of ARVC-associated desmosomal gene analysis. Finally, our retrospective report includes samples from subjects who died more than 15 years ago, without any further family information, which does not enable us to clarify a possible genetic origin of the disease.
Our study confirms a suggested hypothesis about correlation between the presence of inflammatory cell infiltrates, mainly in RV, and a worst pathological heart involvement, including the fibrofatty replacement extension into LV. Our study also strongly suggests that inflammation plays a key role in the severity of the phenotype and disease outcome. However, further research must be performed to provide insights into the inflammatory process and its association with the development of cardiac arrhythmias and SCD.
Take Home Messages
The presence of inflammatory infiltrates in ARVC samples is associated with severe structural heart changes, indicating that inflammation is a modulator of severity in ARVC.
This work was supported by Consejo Superior de Deportes/Ministerio de la Presidencia (003/UPB10/11), CNIC-Translational 2008 (CNIC-03-2008), Eugenio Rodriguez Pascual Foundation, Obra Social ‘la Caixa’ and Montreal Heart Institute Foundation.
Competing interests None.
Patient consent Obtained.
Ethics approval Montreal Heart Institute, University Montreal (Canada).
Provenance and peer review Not commissioned; externally peer reviewed.