Author + information
- Bastiaan J. Boukens, PhDa,b,
- Vichai Benjacholamas, MDc,
- Shirley van Amersfoort, MScb,
- Veronique M. Meijborg, PHDb,
- Cees Schumacher, MScb,
- Bjarke Jensen, PhDa,
- Michel Haissaguerre, MD, PhDd,
- Arthur Wilde, MD, PhDe,
- Somchai Prechawat, MD, MSCf,
- Anurut Huntrakul, MD, MSCf,
- Koonlawee Nademanee, MDf,g,∗ and
- Ruben Coronel, MD, PhDb,d,∗∗ ()
- aDepartment of Medical Biology, Amsterdam Cardiovascular Sciences, University of Amsterdam, Amsterdam University Medical Centers, Amsterdam, the Netherlands
- bDepartment of Experimental Cardiology, Amsterdam Cardiovascular Sciences, University of Amsterdam, Amsterdam University Medical Centers, Amsterdam, the Netherlands
- cDivision of Cardiothoracic Surgery, Department of Surgery, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
- dIHU Liryc, Electrophysiology and Heart Modeling Institute, Fondation Bordeaux Université, Pessac-Bordeaux, France
- eDepartment of Cardiology, Amsterdam Cardiovascular Sciences, University of Amsterdam, Amsterdam University Medical Centers, Amsterdam, the Netherlands
- fDivision of Cardiovascular Medicine, Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
- gCardiac Center, King Chulalongkorn Memorial Hospital, Bangkok, Thailand
- ↵∗Address for correspondence:
Dr. Ruben Coronel, Heart Center, Department of Clinical and Experimental Cardiology, Amsterdam Cardiovascular Sciences, University of Amsterdam, Amsterdam UMC, Meibergdreef 9, PO Box 22660, 1100 DD, Amsterdam, the Netherlands.
Objectives The aim of this study was to investigate the mechanism underlying QRS-slurring in a patient with the early repolarization pattern in the electrocardiogram (ECG) and ventricular fibrillation (VF) storms.
Background The early repolarization pattern refers to abnormal ending of the QRS complex in subjects with structurally normal hearts and has been associated with VF.
Methods We studied a patient with slurring of the QRS complex in leads II, III, and aVF of the ECG and recurrent episodes of VF. Echocardiographic and imaging studies did not reveal any abnormalities. Endocardial mapping was normal but subxyphoidal epicardial access was not possible. Open chest epicardial mapping was performed.
Results Mapping showed that the inferior right ventricular free wall activated the latest with local J-waves in unipolar electrograms. The last moment of epicardial activation concurred with QRS-slurring in the ECG whereas the J-waves in the local unipolar electrograms occurred in the ST-segment of the ECG. Myocardial biopsies obtained from the late activated tissue showed severe fibrofatty alterations in the inferior right ventricular wall where fractionation and local J-waves were present. After ablation, the early repolarization pattern in the ECG disappeared and arrhythmias have been absent since (follow-up 18 months).
Conclusions In this patient, the electrocardiographic early repolarization pattern was caused by late activation due to structurally abnormal myocardium. The late activated areas were marked by J-waves in local electrograms. Ablation of these regions prevented arrhythmia recurrence and normalized the ECG.
The early repolarization pattern refers to slurring or notching at the end of the QRS complex and is associated with sudden cardiac arrest due to ventricular fibrillation (VF) in patients with structurally normal hearts (1,2). In patients with documented cardiac arrest and the early repolarization pattern, implantation of an automated implantable cardioverter-defibrillator (ICD) is the recommended therapy. Identification of patients with the early repolarization pattern at risk for arrhythmias is challenging because it is common in the healthy general population. Previous syncope or a family history of sudden cardiac death have been identified as risk factors but their predictive power is low (1). Recent reports indicate that ablation of myocardium with fractionated potentials can prevent VF in a subset of patients with the early repolarization pattern (3,4). This implies that a localized and potentially arrhythmogenic substrate underlies life-threatening arrhythmias in patients with the early repolarization pattern.
We present the first open chest epicardial electrophysiological mapping of a patient with the early repolarization pattern and VF storms in whom myocardial biopsies were obtained from tissue with fractionated potentials. Our data provide electrophysiological and histological evidence that localized myocardial structural abnormalities can be the underlying arrhythmogenic substrate for life-threatening ventricular arrhythmias and early repolarization in the electrocardiogram (ECG). Epicardial ablation of the structurally abnormal myocardium prevented recurrence of VF and normalized the ECG. We propose that in similar patients with the early repolarization pattern in whom a structural substrate is present, ablation may be the therapy of choice.
This study was performed after written informed consent of the patient, and was executed in accordance with the declaration of Helsinki. The patient underwent standard medical diagnostics and care, and standard resuscitation procedures were followed.
First, the patient underwent an electrophysiologic study involving endocardial mapping using CARTO-mapping system (Biosense Webster Inc., Diamond Bar, California). Subsequently, electrocardiographic imaging (ECGI) mapping was executed with the CardioInsight system (Medtronic, St. Paul, Minnesota). The ECGI methodology has been described before (4,5). Body surface ECG recordings were acquired with a 252-electrode vest wrapped around the torso. Low-dose computed tomography was used to localize the vest electrode positions relative to the heart and the torso. The vest remained in the same position during the electrophysiologic study and programmed stimulation. Body surface ECG recordings from the vest electrodes were acquired before and during the invasive mapping procedure. A 3-dimensional (3D) model of the heart was created using dedicated software (CardioInsight; Medtronic). As a result of pericardial adhesions, it was not possible to gain access to the pericardial space for detailed invasive mapping of the epicardium.
The CardioInsight system automatically displays epicardial wave front patterns on the 3D reconstruction of the patient’s heart based on the solution of the inverse problem (5). In this manner, activation mapping was performed during spontaneous or induced VF. After adequate filtering and phase mapping, dynamic wave front propagation maps were generated. Cardioversion was performed if VF lasted >10 s.
The wave front maps display the electrical wave front at the pi/2 phase value of each ECGI-calculated unipolar electrogram morphology, serving as a surrogate for local activation. We analyzed the VF maps during an initial organized period of VF (the initial 5 s), as previously described (6). VF drivers were defined as either focal breakthrough or full re-entrant activity with a high activation frequency. Focal breakthroughs are detected when centrifugal activation originated from a given site. Rotations are detected when the rotational core, or singularity point, of a rotating wave front is within a 2.5-cm area for ≥1.5 rotations. We then created spatiotemporal density maps displaying the number, location, and spatial extent (of re-entry trajectories) of VF drivers. We marked (colored hexagons) the number and location of epicardial focal breakthrough.
Open chest multielectrode mapping
A midsternal sternotomy was performed. After opening of the pericardium and careful dissection of the pericardial adhesions, sequential unipolar epicardial mapping was executed using a rectangular 8 × 8 multielectrode grid (64 electrodes, 5-mm interelectrode distance). The electrode grid was sequentially placed on 9 locations of the epicardium to cover the entire surface of the heart. The reference electrode was placed in the thoracic wound. A 3-lead ECG was recorded simultaneously. During the entire mapping procedure, a train of 8 S1 stimuli were delivered from the right atrium (cycle length 800 ms), followed by a single S2 from the left ventricular apex (coupling interval 580). Primary data analysis was performed in the operating theatre to guide ablation. Local moments of repolarization were defined as maximum positive dV/dt of the local T-wave (7).
Ventricular epicardial ablation was performed by radiofrequency energy (20–45 W, duration ranging 10 to 60 s) delivered by a ThermoCool catheter (Biosense Webster).
Histology and 3D reconstruction
Biopsies were obtained from selected sites in the right ventricular inferior wall and basal free wall and directly snap frozen in liquid nitrogen. The biopsies were cryo-sectioned (8 μm) and stained with picrosirius red. The 3D model was generated in Amira (version 6.5, Thermo Fischer Scientific, Waltham, Massachusetts). We imported the images (2,560 × 1,920 pixels at 600 dpi, resolution of 1.95 μm/pixel) of 49 histological sections (collagen is red, myocardium yellow) and a 3D reconstruction was generated as published previously (8). The images were imported and converted to grayscale images, which rendered collagen dark gray and myocardium light gray, with a high contrast between the two. We then automatically aligned the images. Alignment errors were corrected manually. Next, we surveyed the image stack for clusters of myocardium, the core of which could be reliably traced on at least 10 consecutive sections (on every section this core of myocardium also had peripheral parts which could not always be traced on the neighboring sections). Myocardial clusters were labeled green.
Diagnosis of the early repolarization pattern
The patient, a 36-year-old man, suffered aborted sudden cardiac arrest at age 30 years and was transferred to the nearest hospital after cardiopulmonary resuscitation in 2012. During the hospital admission, the patient had documented VF and was defibrillated by a DC shock. An ICD was then implanted. VF episodes recurred 1 year later, increasing in frequency from 1 to 4 episodes per month. The patient then received oral Propanolol (10 mg every 8 h). In May 2018, the patient experienced palpitations and dizziness while sitting at home, 10 s before the ICD delivered a shock. Four more appropriate shocks were delivered at the same day and the patient was referred to our hospital.
We observed an apparently healthy young man with a midline sternotomy scar resulting from prior chest surgery performed because of a chest trauma when he was 6 years old. The ECG showed slurring of the QRS complex (≥1 mV) in leads II, III, and aVF, typical of the early repolarization pattern (Figure 1A). An ECG recorded during childhood was not available. Intravenous bolus administration of Ajmaline (1 mg/kg; total 60 mg) did not provoke right precordial ST-segment elevation, a characteristic of the Brugada ECG pattern. Echocardiographic and imaging studies did not reveal any abnormalities. Family screening revealed that an uncle had (unspecified) heart disease. The patient’s parents, his 3 brothers, and his 2 children did not have a history of arrhythmias. The patient’s symptoms and unexplained cardiac arrest meet the standards for diagnosis of a malignant early repolarization pattern according to the Shanghai Score System (9).
An electrophysiological percutaneous catheter study did not reveal endocardial electrical abnormalities (low voltage or fractionated potentials, Supplemental Figure 1). During the electrophysiological study, spontaneous short-coupled premature ventricular contractions and polymorphic ventricular tachycardias occurred frequently (Figure 2A). Subxyphoidal epicardial access was impossible due to pericardial adhesions. Noninvasive ECGI during sinus rhythm and of ventricular fibrillation suggested an arrhythmogenic substrate at the right ventricular free wall (RVFW) epicardium (Figures 2B and 2C).
The patient then underwent open chest surgery to identify a possible epicardial arrhythmogenic substrate and to perform epicardial ablation, if necessary.
Late activation of the right ventricular inferior wall
The electrode grid was placed over 7 anterior locations (Figure 3A) as well as 2 posterior locations of the heart. Figure 3B shows the activation patterns of 3 subsequent beats recorded from the right ventricular outflow tract (RVOT) and RV inferior wall (red in Figure 3A). Figure 3C shows unipolar electrograms (a to g indicate electrode positions in 3B) and lead II of the ECG. The earliest onset of the body surface ECG was taken as the reference time. During S1 pacing, the inferior RVFW activated last, also relative to all other grid positions (Figure 4A) and later than the RVOT (Figures 3A and 3B). Activation of the inferior RVFW concurred with the QRS-slurring in lead II of the ECG, both during the basic (S1, dotted line) and the premature (S2) activation (Figure 3C). Following S2 (coupling interval 580 ms) activation delay occurred and a drastic reduction in the voltage of sites b, d, f, and g, and fractionated local potentials in the same tissue became evident. Note that following S2, a notch has appeared in lead II, coinciding with last activation in the right ventricular inferior wall, and that the RV inferior wall remained activated late despite the difference in activation patterns between the atrial and ventricular beats (see QRS-morphology in lead II).
Early repolarization did not coincide with the early repolarization pattern in the ECG
Based on local multielectrode recordings from 9 different locations, we reconstructed the activation and repolarization patterns of the entire epicardial surface of the heart during atrial pacing (Figure 4A, only anterior view shown). The anterior wall of the right ventricle activated the earliest whereas the right inferior wall activated last, slightly later than the RVOT (85 vs. 83 ms, respectively). Repolarization was the earliest in the anterior wall of the right ventricle. The RVOT and apex of the left ventricle repolarized last. Repolarization in the right ventricular inferior wall was heterogeneous; parts repolarized relatively early (location a) and other parts relatively late (location b). However, the earliest repolarization times occurred much later than the slurring of the ECG (Figure 4B).
The last activated myocardium in the inferior RVFW showed local J-waves in the unipolar electrograms which occurred during the ST-segment of lead aVF, but after the moment of slurring or of the notch in the QRS complex (Figure 4B). Figure 4C (left) shows that the largest J-waves in the local electrograms occurred in the last activated tissue. Indeed, the amplitude and timing of the peak of the J-wave in the local electrograms correlated with the moment of local activation (Figure 4C, middle and right).
Epicardial ablation of structural abnormal right ventricular inferior wall
Two transmural right ventricular biopsies were obtained before epicardial ablation was performed. Fractionation at these sites was confirmed by unipolar and bipolar electrogram recordings obtained with the ablation electrode (Figure 5A). One biopsy was taken from the last activated tissue of the inferior RVFW, showing local J-waves or fractionation (Figure 5A), and 1 from the basal right ventricular wall. The biopsy of the right ventricular inferior wall showed extensive fibrosis (picrosirius red staining, collagen is red) with fragments of surviving myocardium. In contrast, the basal right ventricle only showed increased diffuse fibrosis without loss of myocardium (Figure 5B). Three-dimensional reconstruction of the fibrotic tissue showed a fine network of interconnecting individual bundles of myocardium (Figure 5C, green).
Immediately after radiofrequency ablation of the epicardial myocardium that showed fractionated potentials and/or J-waves, the QRS notching in the ECG leads was absent (not shown). The early repolarization pattern has remained absent during the follow-up period (publication date) and the patient has been free of VF since the procedure (Figure 1B).
In this case study, we combined high-resolution epicardial electrophysiological mapping and histological analysis in a patient with the documented early repolarization pattern. Although endocardial abnormalities were not detected, we show that localized structural abnormalities in the subepicardium were the cause for the electrocardiographic pattern and for VF storms. Ablation of these regions with structural abnormalities normalized the electrocardiogram and prevented arrhythmia recurrence.
We describe severe but localized interstitial fibrosis in RV myocardium and recorded fractionated unipolar electrograms from the same locations. This resembles post-infarcted myocardium, where surviving myocardial strands within the fibrotic infarcted tissue constitute a substrate for life-threatening arrhythmias (10). The loose myocardial network within a fibrotic area may well form the basis for delayed conduction or failure of conduction by current-to-load mismatch (11,12). If local fibrosis is present in already late activated myocardium, as is the case for the subepicardium of the right and left ventricular wall of the described patient, local activation may be delayed beyond the end of the QRS complex and become visible as slurring or notching. In this patient, the structural abnormalities provided the substrate for arrhythmias as confirmed by ablation of these sites that resulted in an arrhythmia-free follow-up.
Our data also indicate that a loose myocardial network within a fibrotic area is associated with local J-waves. J-waves in unipolar electrograms are caused phase-1 repolarization at the action potential level, particularly at the subepicardium. The common mechanistic explanation of the early repolarization pattern is that enhanced phase-1 repolarization causes J-waves in local electrograms, which are visible in the ECG as notching or slurring of the QRS complex (9). However, in our patient, local J-waves recorded in late activated myocardium did not coincide with the QRS-slurring and were not visible in the ECG. This observation challenges the common view that all cases of early repolarization pattern are caused by enhanced phase-1 repolarization.
Recent reports have shown that idiopathic VF is associated with fractionated local unipolar electrograms or with electrograms with a J-wave (3,13). Our patient had both local fractionated potentials and local J-waves, indicating that these conditions can coexist. We have previously argued that local J-waves can be recorded at epicardial sites that are activated late (14). The plots in Figure 3C confirm that both the time and the amplitude of local J-waves correlated with activation time. Our data show that the choice of potential ablation sites can be guided by local fractionation activity and local J-waves, as expression of late activation (14).
We show that the inferior wall activated late and repolarized relatively early compared to the surrounding myocardium (Figure 3A). Consequently, the activation-recovery-interval, surrogate for action potential duration, was short in this region. Short activation-recovery-intervals in the right inferior wall have been described before in patients with a similar early repolarization pattern in the ECG, as the case we describe here (15). However, it is unlikely that the brief action potentials alone explain the early repolarization pattern in the ECG as they are only present in regions that are activated late. Also, in our patient, the earliest moment of repolarization occurs much later than the moment of QRS-slurring in the ECG. The mechanism of the relatively early repolarization in this late activation myocardium is unclear. We speculate that adjacent unexcited (or not yet activated) myocardium electrotonically shortens local repolarization. Our observations do not allow speculation about the transmural repolarization gradient.
We cannot exclude the possibility that the chest surgery that the patient had undergone in his youth has contributed to the localized structural changes. Neither can be ruled out the possibility that the antiarrhythmic effects were mediated by a suppression of the triggering mechanism associated with the scar tissue. A common view is that a J-wave syndrome is only truly a J-wave syndrome by exclusion of a scar or other structural abnormalities. Our observations, however, show that small structural abnormalities can be present in patients diagnosed with the early repolarization syndrome. In our patient, the diagnosis of early repolarization syndrome (or early repolarization pattern) was made based on aborted sudden cardiac death, documented VF, or polymorphic tachycardia in combination with the early repolarization pattern in the ECG (Shanghai score system) (9). The patient does not meet the criteria for any pathology listed as other causes for early repolarization pattern. This means that current (clinical) detection methods may not be sufficient to exclude small regional structural abnormalities that can underlie VF in patients with the early repolarization syndrome. This makes the mechanistic insight into the early repolarization syndrome provided by this study relevant for patient management.
We show that the early repolarization pattern can result from localized structural abnormalities (Central Illustration). These abnormalities were diffuse and not detected with standard clinical tools. We do not imply that structural abnormalities are present in all patients with the early repolarization pattern. However, our observation does challenge the definition of a structurally normal heart and the notion that a single functional pathophysiological mechanism can be established for all patients with the early repolarization pattern.
COMPETENCY IN MEDICAL KNOWLEDGE: The current view is that patients diagnosed with the early repolarization syndrome have structurally normal hearts. We demonstrate that clinically concealed right ventricular subepicardial structural abnormalities cause ventricular fibrillation in a patient with the early repolarization syndrome.
TRANSLATIONAL OUTLOOK: Localized structural abnormalities (fibrosis) can cause the early repolarization pattern and can form an arrhythmogenic substrate. Detailed electrophysiological and anatomic imaging should be performed to identify a substrate that can be ablated in patients with the early repolarization syndrome.
The authors thank Qing Lou for assisting with the analysis of the ECGI data.
↵∗ Drs. Nademanee and Coronel contributed equally to this work.
This study was supported by the Leducq foundation (16CVD02 RHYTHM, to Dr. Coronel). Dr. Boukens received funding from the Dutch Heart Foundation (2016T047). The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All other authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- electrocardiographic imaging
- implantable cardioverter-defibrillator
- right ventricular free wall
- ventricular outflow tract
- ventricular fibrillation
- Received April 9, 2020.
- Revision received June 12, 2020.
- Accepted June 16, 2020.
- 2020 The Authors
- Haïssaguerre M.,
- Nademanee K.,
- Hocini M.,
- et al.
- Nademanee K.,
- Haissaguerre M.,
- Hocini M.,
- et al.
- Nash M.P.,
- Mourad A.,
- Clayton R.H.,
- et al.
- Kawara T.,
- Derksen R.,
- De Groot J.R.,
- et al.
- Rohr S.,
- Kucera J.P.,
- Fast V.G.,
- Kleber A.G.
- Boukens B.J.,
- Opthof T.,
- Coronel R.