|Home | About | Journals | Submit | Contact Us | Français|
Parietal block, defined as intra right ventricular (RV) conduction slowing is a major diagnostic criterion for ARVD/C.
We evaluated the utility of total RV endocardial activation duration (EAD) measured by 3D electroanatomic mapping during sinus rhythm in the diagnosis of ARVD/C.
25 consecutive patients with frequent LBBB morphology PVCs who underwent electroanatomic mapping as a part of the evaluation for ARVD/C were included in the study. All patients were evaluated using standard protocol that included, ECG, signal averaged ECG, Holter, Echocardiography, and MRI. Invasive testing was performed as indicated. Total RV EAD was measured as the time interval between the onset of RV activation to the latest activated region in the RV.
Mean age of the study subjects was 38±11 and 32% were men. 14 subjects were diagnosed as ARVD/C using task force criteria and the remainder had idiopathic VT. While the surface QRS durations were similar, the total RV EAD was significantly prolonged in ARVD/C compared with idiopathic VT (83.9±10 msec vs. 50.8±7 msec, p<0.001). None of the idiopathic VT subjects had RV EAD of > 65 msec. RV EAD also showed significant negative correlation with RV ejection fraction.
Total RV EAD obtained by 3D electroanatomic mapping is a sensitive marker of intra RV conduction delay in ARVD/C and a total RV EAD of >65 msec accurately differentiates ARVD/C from idiopathic VT.
Arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C) is an inherited cardiomyopathy characterized clinically by ventricular arrhythmias and histologically by fibrofatty replacement of the right ventricle (RV)1,2,3. The diagnosis of ARVD/C is often challenging and is based on a set of major and minor criteria encompassing structural, electrocardiographic, and histological criteria proposed by the Task Force of the Working Group on Cardiomyopathies in 19944. Among the many ECG parameters that have been associated with ARVD/C, the most important are those that are markers of delayed activation of the RV. These include right precordial QRS prolongation, delayed “S” wave upstroke and terminal activation delay5,6,7,8. The presence of delayed potentials on signal averaging also has been attributed to this delayed activation of the RV.
Until recently, delayed RV activation in ARVD/C was attributed to slow electrical conduction through areas of fibrofatty tissue in the RV. However, the past five years have seen remarkable advances in the understanding of the pathophysiologic basis of ARVD/C. It is now recognized that typical ARVD/C is caused by mutations encoding desmosomal proteins that maintain cell to cell adhesion9,10,11. We and other investigators have reported significant gap junction remodeling with failure of localization of connexin 43 at the gap junctions in ARVD/C9,12,13,14. Slowed electrical conduction within the RV was also demonstrated in Plakoglobin knock out mice despite the lack of histological evidence of ARVD/C15. Based on the cumulative clinical observations over the past 20 years and the new science of ARVD/C, it has become clear that activation delay is a hallmark of ARVC/D. Accordingly, the present study tested the hypothesis that the timing and pattern of activation of the RV measured by direct 3D electroanatomic mapping may be an early and a sensitive marker for ARVC/D. We investigated the role of activation, rather than voltage mapping, in the diagnosis of patients suspected of having ARVC/D.
The study population included 28 consecutive patients who underwent electrophysiologic study including 3D electroanatomic mapping as a part of their evaluation for ARVD/C at the Johns Hopkins ARVD/C center. Each of the patients had a history of PVCs or non-sustained VT with LBBB morphology. All the patients were evaluated using a standard protocol that included a 12 lead ECG, Signal Averaged ECG, Holter, 2D echocardiography and MR imaging. All patients underwent 3D electroanatomic mapping according to a standard protocol. First degree relatives of patients with ARVD/C were enrolled if they had at least one minor criterion for the diagnosis of ARVD/C namely precordial T wave inversions beyond V2, abnormal filtered QRS duration on signal averaged ECG, > 200 premature ventricular contractions on Holter monitoring and minor structural abnormalities of the RV detected on echocardiography or magnetic resonance (MR) imaging. Echocardiography and MR imaging were performed according to a standard protocol16,17. Endomyocardial biopsy (EMB) was performed in 7 patients at the discretion of the evaluating physician. Informed consent was obtained in all patients and the study was approved by the institutional review board.
All patients underwent high density 3D electroanatomic voltage mapping of the right ventricle using the CARTO system (Biosense-Webster) during sinus rhythm. A minimum of 145 points were sampled throughout the RV and RVOT. A 7F Navi-Star catheter, which consisted of a 4-mm distal tip electrode and a 2-mm ring electrode with an interelectrode distance of 1 mm, was introduced into the RV under fluoroscopic guidance and used as the mapping catheter. The catheter was placed at multiple sites on the endocardial surface to record bipolar electrograms from RV inflow, anterior free wall, apex, and outflow. The bipolar signals were filtered at 10 to 400 Hz and were displayed at 100-mm/s speeds on the CARTO system. Bipolar electrograms were analyzed with regard to amplitude, duration, relation to the surface QRS, and presence of multiple components. A recording was accepted and integrated in the map when the variability in cycle length, local activation time stability, and maximum beat-to-beat difference of the location of the catheter were <2%, <3 ms, and <4 mm, respectively. In addition, adequate catheter contact was confirmed by concordant catheter tip motion with the cardiac silhouettes on fluoroscopy and provoked PVCs due to contact. The peak-to-peak signal amplitude of the bipolar electrogram was measured automatically. A 3D geometry of the RV chamber depicting the local activation times of the bipolar electrograms recorded at each site was constructed in real time with the electrophysiological information color coded and superimposed on the reconstruction. Mapping points were acquired until a complete electroanatomic map of the RV had been generated. Electrogram amplitude used to define normal RV endocardium was set at 1.5 mV and electroanatomic scar area was defined as an area 1 cm2 including at least 3 adjacent points with bipolar signal amplitude <0.5 mV18. Local activation was defined as the first positive deflection in the bipolar electrogram. For fractionated electrograms with multiple components the first positive component was taken as the local activation time. Total right ventricular endocardial activation duration (RV EAD) was measured as the time interval from the onset of RV endocardial activation to the latest RV endocardial activation recorded during sinus rhythm.
Data are represented as mean±SD. Comparisons between the ARVD/C and idiopathic VT were performed with unpaired t tests (Intercooled STATA 7, Chicago, Illinois) and correlation between variables was tested using correlation coefficients. A p vaue of <0.05 was considered significant.
Fourteen of the 28 patients were diagnosed with ARVC/D/C based on the task force criteria. 11 patients had no structural abnormalities detected both during the initial screening and the electrophysiologic testing and were diagnosed with idiopathic VT. Of the remaining 3 patients, one was diagnosed with probable ARVD/C as he met partial task force criteria. Two patients had idiopathic cardiomyopathy with biventricular dysfunction. These three patients were excluded from further analyses.
Baseline characteristics of the 25 study group patients are provided in Table 1. Their mean age was 38± 11 yrs and 36% (9/25) were men. The majority had palpitations as the presenting symptom. Five of the 14 ARVD/C were asymptomatic at the time of their evaluation. One ARVD/C patient was on Sotalol, which was discontinued 3 days prior to the electrophysiologic testing. Six of the 14 had normal RV systolic function (RVEF>50%) and each of these patients met criteria based on the positive family history and two minor criteria. Average RVEF was 49% and only two of the 14 patients had severe functional abnormalities on MR imaging (RVEF<40%). Mean RVEF however was significantly lower in ARVD/C compared to idiopathic VT patients (49±11 vs. 57±5, p=0.002). The mean QRS duration was not significantly different between patients with or without ARVD/C, and none of the subjects had complete RBBB pattern. No complications were observed either during EP study or during EMB.
All patients were in sinus rhythm during CARTO mapping, and none had sinus node or AV node dysfunction. The mean number of sites sampled in RV electroanatomic mapping was 145±21, with an average mapping period of 32±8 minutes. Three ARVD/C patients and all of the idiopathic VT patients had a completely normal voltage map. The remaining ARVD/C patients demonstrated one or more areas of endocardial low voltage mainly confined to the RV outflow tract and the basal perivalvular region of the RV Table 2. The RV anterior free wall and septum had preserved bipolar electrogram amplitude (4.4±0.7 mV) and duration (37.2±0.9) in 12/14 ARVD/C patients. None of the idiopathic VT patients had either global or regional functional abnormalities on MRI. No significant correlation was observed between MRI abnormalities and RV EAD (Table 2).
Figure 1 displays a representative voltage (left) and activation map from a patient with ARVD/C. Postero-anterior view of the voltage map shows low voltage areas in the peri valvular region. An activation map during sinus rhythm reveals delayed activation of the RV outflow tract with preserved endocardial voltage. Local endocardial activation of the RV outflow is occurring after the end of the QRS complex, a finding consistent with delayed activation.
Table 3 provides the electrophysiologic data including the activation times recorded in the “Triangle of dysplasia” for each of the ARVD/C patients and idiopathic VT subjects. Time to onset of RV endocardial activation was similar between ARVD/C and idiopathic VT (6± 8 vs. 8±5, P = NS). Earliest RV endocardial activation occurred in the anterior RV in the majority of the patients consistent with prior reports. RV outflow endocardial activation was the latest RV structure to be activated in the majority of patients (18/25, 72%). The basal subtricuspid region of the RV was the latest activated structure in the remainder of the patients. QRS duration was not significantly different between the two groups; however, the total RV endocardial activation duration was significantly prolonged in ARVD/C compared to idiopathic VT (83.9±10 msec vs. 50.8±7 msec, p<0.001) (Figure 2). None of the patients with idiopathic VT had a total RV EAD > 65 msec. Thus an RV EAD of >65 msec had high positive predictive value (100%) for the diagnosis of ARVD/C.
Total RV EAD obtained by electroanatomic mapping showed better correlation with RVEF (R2=0.76) as compared to surface QRS duration (R2=0.56), however this was not statistically significant.
Comparison various surface ECG markers of ARVD/C with the RV endocardial activation duration is shown in Table 3. Among the ECG parameters, T wave inversion alone had the best sensitivity (64%) which was significantly better that “S wave upstroke” or Terminal Activation Delay” (36% and 50% respectively, p<0.01 for both). Right precordial QRS duration showed a modest correlation with total RV EAD with an R2 of 0.52. None of the other surface ECG parameters of activation delay showed a good correlation with total RV EAD.
We report for the first time the utility of electroanatomic RV endocardial activation duration to evaluate intraventricular conduction in ARVD/C. Surface QRS derived measures of RV activation delay may underestimate the degree of conduction delay in ARVD/C. We also have demonstrated that this novel parameter can be used in the clinical setting to differentiate ARVD/C from idiopathic VT. Endocardial activation duration is abnormal in ARVD/C even in the absence of overt RV dysfunction and hence may be useful in concealed ARVD/C. And finally, RV EAD shows excellent correlation with the degree of RV dysfunction.
In the past several years, a number of studies have reported that 3D electroanatomic mapping may be of value in the diagnosis of ARVC/D by identifying areas of low voltage suggestive of myocardial scar. Perivalvular and outflow tract scarring have been reported in ARVD/C, although this finding is not unique to ARVD/C and is shared by other non-ischemic cardiomyopathies 19. Other studies have reported that endomyocardial biopsy of these low voltage areas is associated with fibrofatty replacement. Electroanatomic mapping has also be used to facilitate catheter ablation of VT in patients with ARVC/D/C 18,19,20. The majority of patients in these prior studies had global RV dysfunction and patients in early stages of disease and patients with idiopathic VT were not included. Another important limitation of voltage mapping is that there is no way to distinguish low voltage resulting from poor tissue contact with low voltage resulting from scar and fibrosis. This problem is particularly important when mapping the perivalvular region as catheter contact with the myocardium in this region is extremely difficult.
Despite the encouraging reports of prior studies that have reported that voltage mapping is of value in the assessment of patients suspected to have ARVC/D/C, our experience has not been as positive. In fact, during the past decade we have seen a large number of patients who have been diagnosed with ARVC/D/C based on low voltage identified on a CARTO map. In the great majority of these patients, a complete screening for ARVC/D revealed no other evidence of the disease. This low specificity of low voltage areas in the RV for detecting ARVC/D likely reflects the fact that low voltage can either reflect myocardial scar and fibrosis or it can reflect poor tissue contact. It is striking that although voltage mapping has been explored as a diagnostic tool, investigators have completely ignored the information on activation. This is particularly notable since an activation map can be obtained simultaneously with a voltage map. Further local activation time measurement does not share the limitations of voltage mapping with respect to catheter contact.
Durrer et al 21 originally reported the time sequence of RV activation in an isolated perfused human heart in 1970. RV endocardial activation, mediated by rapid conduction via purkinje fibers, starts 5-10 msec after the onset of the intracavitary potential at the base of the anterior papillary muscle and is usually completed by 45-55 msec21. Epicardial breakthrough often occurs on the RV anterior wall and radially spreads to the RV base and the RV outflow tract, with these locations being the latest activated regions in the RV. In all the isolated hearts in that study RV activation (both endocardial and epicardial) was completed before the end of the QRS complex. Under physiologic conditions, local RV epicardial activation is completed 12-20 msec before the end of the QRS complex and RV activation beyond the offset of the QRS complex is considered pathological21. In 8/14 patients with ARVD/C in our study RV endocardial activation continued after the end of the QRS complex, a phenomenon that was observed in none of the non-ARVD/C patients. These delayed regions were not always associated with decreased endocardial voltage and/or fractionated potentials. Thus, it is hard to argue that the delay was secondary to local fibrosis although microscopic fibrosis in the absence of decreased voltage cannot be completely excluded.
Intra right ventricular conduction delay termed “parietal block” represents a major criterion for the diagnosis of ARVD/C4,6. Several studies have reported 55-100% sensitivity and a high specificity of this parameter. Other surface ECG correlates of slowed or delayed RV conduction which have been proposed as criteria for ARVC/D include a d “S” wave upstroke, terminal activation delay, and a ratio of the right and left QRS duration>1.16,7,8 Despite the ease of use, these parameters are at best surrogates of RV activation and are limited by lead positioning and error of measurement. These non-invasive parameters have never been validated against direct measurement of RV activation and the correlation between the two has been presumptive. Our study is the first to compare true activation of the RV with surface ECG parameters. The results of this study that the correlation is not optimal in ARVD/C. Surface ECG measures of RV activation delay tends to underestimate the delayed conduction and hence may not be sensitive to early forms of disease. T wave inversions beyond V2 on the other hand had a better sensitivity for ARVD/C diagnosis compared with the “S” wave upstroke and Terminal Activation Delay. This possibly reflects a higher proportion of patients with early ARVD/C in our study, with 6/14 patients having a normal RV global function.
Our study has important clinical implications. Electroanatomic voltage mapping is now routinely performed in patients with suspected ARVD/C following the reports suggesting its utility in diagnosis and risk stratification. The same data can be used to obtain total endocardial activation time. Based on the results of our study we would propose that a total RV EAD of >65 msec should raise the suspicion of ARVD/C despite normal voltage mapping. The region of latest activation is often the RV outflow tract or the sub-tricuspid region, hence limited RV mapping of these regions may suffice for diagnosis although this remains to be prospectively evaluated. Delayed RV activation is demonstrable even in the absence of global functional abnormalities and RV endocardial scar and may have an important role in early diagnosis of ARVD/C. RV epicardial mapping was not performed in our study; however we do not feel that this is a major limitation as this is not clinically performed in the routine evaluation and requires expertise that is not widely available.
RV endocardial activation duration derived by electroanatomic mapping provides a direct measurement of conduction delay in ARVD/C. This novel parameter appears to be high positive predictive value and yet differentiates ARVD/C from idiopathic VT. The total RV EAD shows excellent negative correlation with RVEF. Further studies are required to validate this parameter in early diagnosis of ARVD/C.
The authors wish to acknowledge funding from the National Heart, Lung, and Blood Institute (K23HL093350 to HT). The Johns Hopkins ARVD Program is supported by the Bogle Foundation, St. Jude Medical Foundation, the Healing Hearts foundation, the Campanella family, and the Wilmerding Endowments. We are grateful to the ARVD patients and families who have made this work possible.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Harikrishna Tandri, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Angeliki Asimaki, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts, USA.
Theodore Abraham, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Darshan Dalal, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Laurens Tops, Department of Cardiology, Leiden University Medical Center, Leiden, the Netherlands.
Rahul Jain, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Jeffrey E Saffitz, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts, USA.
Daniel P Judge, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Stuart D Russell, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Marc Halushka, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
David A Bluemke, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
David A Kass, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Hugh Calkins, Division of Cardiology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA.