Historical Background
From the First Description to Current Views
The term idiopathic comes from Greek roots idios and pathos, meaning a disease that does not arise from any known cause. The first use of ‘idiopathic ventricular fibrillation’ was made by Ledwich and Fay in 1969 in a report on a patient in whom cardiac examination, 12-lead ECG and chest X-ray were normal (Figure 1).1
Over time, the proportion of patients labelled as having idiopathic VF has progressively decreased, due to improvements in the diagnostic workup of patients with resuscitated sudden cardiac death. Coronary catheterisation and echocardiography in the late 1960s were the first important diagnostic tests that markedly improved the definition of ‘apparently normal heart’. More recently, major progress has been made with cardiac MRI and pharmacological challenges. Arrhythmogenic cardiomyopathy and electrical diseases represent a significant proportion of diagnoses attributed during follow-up in patients with an initial diagnosis of idiopathic VF. In parallel, genetic testing and improved understanding and recognition of cardiac channelopathies (long QT syndrome, short QT syndrome, catecholaminergic polymorphic ventricular tachycardia [CPVT], calcium release deficiency syndrome [CRDS]) and arrhythmic syndromes (Brugada, early repolarisation) have further restricted the span of idiopathic VF.2,3
Therefore, idiopathic VF is a complex clinical entity that is characterised by an evolving scope over time. The main stem is that it is a diagnosis per exclusion: VF occurring in the absence of structural or electrical heart disease, or reversible metabolic and toxicological conditions. Since the 1920s, there have been reports of apparent idiopathic VF, with the first case probably described by William Dock in 1929.4 Next, two patients with apparent idiopathic VF were identified more than 70 years ago and had VF initiated by short-coupled premature ventricular contraction (PVC). Both were male, aged 38 and 37 years and suffered from multiple episodes of palpitations and syncope.5,6 The first case of idiopathic VF storms was reported by Moe in 1948.5 In 1994, Leenhardt et al. described the first series of 14 patients (50% women, mean age 34.6 ± 10 years) with short-coupled torsade de pointes based on the QRS morphology of the arrhythmia.7 The coupling interval of the PVC initiating the arrhythmia was 245 ± 28 ms (range 220–300 ms).
In 2002, we reported mapping and ablation of idiopathic VF related to short-coupled PVCs, demonstrating the involvement of the Purkinje system in this clinical presentation.8 More recently, Steinberg et al. reported the prevalence of short-coupled idiopathic VF in a multicenter study involving 364 patients with idiopathic VF. Selecting a coupling interval <350 ms, they found 24 (6.6%) patients. These patients had a very high risk of sudden cardiac death.9
Even today, patients presenting idiopathic VF are heterogeneous, although subgroups with common features can be determined. We recently proposed that idiopathic VF can be split into two distinct phenotypes, with potential overlap in an individual patient.10 In patients with Purkinje-related idiopathic VF, short-coupled ventricular ectopy can be documented in a significant proportion of patients and ablation of the culprit tissue is a reasonable strategy to avoid recurrence. In patients with microstructural idiopathic VF, localised myocardial alterations can be unveiled by thorough electro-anatomical mapping and are amenable to a substrate-elimination strategy.
From a mechanistic perspective, it is useful to distinguish VF initiation and maintenance. In idiopathic VF, ventricular arrhythmias are frequently polymorphic from the onset, which distinguishes them from VF following ventricular tachycardia (VT). In idiopathic VF patients, the first beat of the arrhythmia originates from the Purkinje system in up to 93% of cases.8 More rarely, initiation may involve the ventricular myocardium, namely the right ventricular outflow tract or the papillary muscles. Purkinje cells are also active during VF maintenance, as shown by Newton et al. and Tabereaux et al.11,12 This could be explained by the resistance of Purkinje cells to prolonged ischaemia.
In our initial report including 27 patients with idiopathic VF, a Purkinje origin was demonstrated in 23 of 27 (93%) patients.8 The culprit tissue was in the left ventricular septum in 10 patients, in the anterior right ventricle in nine patients and in both ventricles in four patients. A second multicenter study including 38 patients showed that the PVC origin was in the Purkinje system in 33 of 38 (87%) patients.13 Purkinje PVCs came from the left, right and both ventricles in 14, 16 and three patients, respectively. A myocardial origin was identified in five patients, the majority being from the right ventricular outflow tract (four of five patients).
Characteristics of arrhythmia recurrences on ICDs provide phenotypic markers of the distinct and hidden substrates underlying idiopathic VF. In a recent study, we reported that among 95 patients with idiopathic VF, microstructural alterations could be identified in 41, and Purkinje-related anomalies in 54.14 A total of 117 arrhythmia recurrences were recorded, 91% of which were VF. Three variables were mostly discriminant. Sinus tachycardia (≤570 ms) was more frequent in microstructural cardiomyopathic alterations (35.9% versus 13.4%, p=0.014) whereas short-coupled (<350 ms) triggers were most frequent in Purkinje-related VF (95.5% versus 23.1%, p=0.001), which also had shorter VF cycle length (182 ± 15 ms versus 215 ± 24 ms, p<0.001). Ectopy was inconsistently present before VF.
Management of idiopathic VF has evolved over time, from the empiric and sometimes successful use of antiarrhythmic drugs to the implantation of automatic defibrillators in the 1980s and ablation in the 2000s, affecting prognosis.
Long-term Prognosis
Since its first description, idiopathic VF remains an uncommon condition as it accounts for <10% of all VF episodes. However, idiopathic VF must be considered after unexplained cardiac arrest in an otherwise healthy individual. It is more frequent in young adults, as up to 35% of cases of sudden death remain unexplained in patients between 18 and 35 years old.15,16
The UCARE registry reported a 30% recurrence of ventricular arrhythmia after 3 years from cardiac arrest in idiopathic VF patients.17 At 10 years of follow-up, a large European registry reported a cumulative event-free survival of 67%.18 In a study by Pannone et al., the authors reported a ventricular arrhythmia-free survival of 65% at 9 years of follow-up.19 The rate of inappropriate shocks is 11.1% (approximately 1.2% per year), which is low.20
Ablation of VF is associated with high rates of acute success and long-term freedom from recurrence. In a multicentric study including 38 idiopathic VF patients who underwent catheter ablation, we reported that 31 of 38 (82%) patients were free from VF recurrence after a mean follow-up of 63 months.13 VF recurrence occurred in the remaining 7 (18%) patients after a median of 4 months. The presence of bundle branch block before ablation was the only parameter associated with worse outcome and with VF recurrence. There was no difference in outcome between patients with Purkinje triggers and those with muscular triggers.
Current State of the Art and Progress
Idiopathic VF Subtypes
Purkinje-related Idiopathic VF
The mechanisms underlying the arrhythmogenic potential of the Purkinje network are imperfectly understood. Cellular specificities of Purkinje cells and their cable-like configuration favour triggered activity, and the network is prone to developing re-entry. Purkinje ectopy is often polymorphic. There is a predominance of right ventricular Purkinje arrhythmias in male patients, and a predominance of left or biventricular Purkinje arrhythmias in female patients.21 Purkinje premature contractions have a narrower QRS duration than muscular ectopy, particularly when they originate from the left Purkinje system (<120 ms) where they exhibit a right bundle branch block morphology.22 There is a right or left deviation when originating from the anterior or posterior Purkinje fibres respectively. Purkinje premature contractions from the right arborisation have a left bundle branch block morphology and a wider QRS duration (>140 ms) with an initially rapid deflection. Short coupling (R on T) PVCs favour Purkinje origin but are also seen in ventricular origin. In a Dutch registry, 31 of 34 (91%) patients with documented VF onset had short-coupled PVC triggering VF. Patients with short-coupled VF showed a higher shock burden and a higher incidence of electrical storm than patients without short-coupled VF.23
Purkinje ectopy may have a long coupling interval (>350 ms).24 Belhassen et al. found that 17.4% of patients with short, coupled VF also had VF episodes initiated by PVCs with a coupling interval >350 ms.25 In a study by Surget et al., long-coupled PVCs were defined as PVCs occurring after the end of the T wave.24 Seventy-nine patients with idiopathic VF were reviewed, and 12 (15.2%) met this definition. All VF episodes occurred at rest. Invasive mapping showed that all PVCs originated from the distal Purkinje system (92% from the left). The mean coupling interval of long-coupled PVCs was 418 ± 46 ms. Long-coupled PVCs can induce VF by themselves or in association with short-coupled PVCs. When both short- and long-coupled PVCs are recorded in an individual they present different morphologies in most cases.
Microstructural Abnormality-related Idiopathic VF
The most frequent abnormality is the presence of localised areas of slow-conducting myocardium. These areas likely indicate microstructural alterations of the myocardium because of fibrosis, fatty or inflammatory infiltration, or cellular pathologies. The location is more often epicardial than endocardial.10
Stepwise Diagnostic Workup and Management
Detailed History
A detailed family history of clinical circumstances, preceding syncope, is taken (Figure 2).
A normal resting ECG is a prerequisite to idiopathic VF. The signal-averaged ECG is also normal. An ECG with high precordial leads should be performed as it may unveil Brugada syndrome.26 The initiation of idiopathic VF recorded on a 12-lead ECG is the cornerstone for the diagnosis and ablation strategy. As described above, the initiation will unveil the mechanism and the idiopathic VF subtype.
The interpretation of spontaneous Purkinje-PVCs occurring without VF is difficult as they are suggestive of the mechanism of sudden cardiac death but are not necessarily causal. Even when Purkinje-related idiopathic VF is confirmed, the culprit PVC(s) may be different from the ones recorded after the event.
An exercise test is useful to unmask subclinical primary electrical disorders (long QT syndrome, CPVT).27
Echocardiography and cardiac MRI are performed to eliminate structural heart disease. Cardiac MRI has been shown to yield 12.5–19% of additional diagnoses (especially arrhythmogenic cardiomyopathy) in patients initially diagnosed with idiopathic VF.28 A cardiac tomography scan is useful to exclude a coronary artery disease.
Pharmacological Tests
Ajmaline Test
Ajmaline testing is arguably the most important pharmacological test in patients with suspected idiopathic VF.29 The goals of the ajmaline test are to exclude Brugada syndrome and document potential short-coupled PVCs.
One-third of Brugada syndrome patients with spontaneously documented VF do not exhibit spontaneous type 1 ECG, but only after ajmaline administration.27
In a multicentre retrospective study, we reported the frequent identification of short-coupled PVCs in idiopathic VF patients, during administration of sodium blockers, with a morphology compatible with a Purkinje origin.30 This phenomenon was reproducible in all nine patients who underwent a repeat test. In four patients, the test was the only way to reveal Purkinje ectopy and the patients received an implantable cardiac defibrillator. A spontaneous VF was then observed in three of the patients after a median follow-up of 28 months.
Epinephrine Test
Epinephrine test is performed to rule out a concealed long QT syndrome. However, false-positive tests are common, and QT variations may be difficult to interpret. Mental stress test appears more specific.31,32
Isoprenaline Test
An isoprenaline test may unveil long QT syndrome, CPVT and arrhythmogenicity in relation to a concealed or overt cardiomyopathy.32 An infusion of 3 minutes at a dose of 45 µg/min produces a mean peak sinus rate of 152 ± 18/min. This test has a higher arrhythmogenic power than exercise testing, inducing non-sustained VTs in 74–85% of arrhythmogenic right ventricular cardiomyopathy and 42% of dilated cardiomyopathies. In contrast, non-sustained VTs are induced in only 2–2.7% of patients with PVCs and normal hearts at echocardiography. In idiopathic VF patients, this test may induce PVCs suggestive of microstructural anomalies.
Genetics and Familial Screening
Pathophysiology and molecular mechanisms of idiopathic VF are not known; however, genetic testing should be performed systematically. Approximately 20% of patients with idiopathic VF have a positive family history of sudden cardiac death, suggesting a genetic contribution in at least a subset of cases.33–35
Genes that have been associated with idiopathic VF are DPP6, CALM1, RYR2, DSP, TTN, FKTN, TRPM4, MYH7, ANK2, CACNA1C, DES and IRX3.
Verheul et al. showed in a retrospective study that genetic testing is performed in 91% of cases, with a 15% yield of likely pathogenic or pathogenic variants. Variants of uncertain significance are found in 30% of cases and 55% of patients do not have any abnormal variants. In total, finding a genetic variant led to a diagnosis in 2% patients.36 Over the years, the use of gene panel testing has increased (62% in idiopathic VF patients before 2010 versus 87% after 2010, p<0.01). In a study by Pannone et al. of a cohort of 45 idiopathic VF probands undergoing genetic analysis with a broad gene panel, the diagnosis yield for pathogenic or likely pathogenic variants was 6.7%.19 Patients with no variant or a variant of unknown significance had higher ventricular arrhythmia-free survival during the follow-up compared with patients with a pathogenic variant. Genetic linkage analysis for idiopathic VF patients has exposed a non-monogenic disorder and, often, multiple genetic variants are identified without a link to VF arrhythmogenesis.
It is recommended (class 1) to perform genetic testing in sudden cardiac arrest survivors with a suspected genetic cause, and testing should only include genes with an evident gene disease association.4 VF may be a preclinical phenotypical expression of cardiomyopathy or arrhythmia syndromes.
Dipeptidyl-aminopeptidase-like protein 6 (DPP6)-specific haplotype has been associated with idiopathic VF, mainly in the Netherlands (founder effect).37 It is present in 10% of cases. The prognosis is poor as 50% of carriers experience sudden cardiac death before the age of 58. A study described that in the presence of DPP6-VF haplotype, Ito currents are increased in Purkinje fibres.
Calmodulin1 (CALM1) regulates calcium channels. Carriers of a specific variant (c. 268 T>C) present VF before adulthood.
Ryanodine receptor (RYR2) mutations have been associated with idiopathic VF, CPVT and CRDS.
Familial screening encompasses resting ECG, exercise testing and echocardiography in first-degree relatives.
Invasive Electrophysiology Study
CRDS is due to a loss-of-function variant in the RYR2 gene. It alters the action potential duration and increases the risk of early afterdepolarisations. All patients have a negative exercise test result. Recent data on 19 patients with CRDS showed that a life-threatening arrhythmia occurred in 50% of patients at 7 years of follow-up. Ni et al. showed a unique repolarisation response on an ECG after provocation with brief tachycardia and a subsequent pause in CRDS cases and mouse models, which is absent from the controls. Brief tachycardia and a subsequent pause (either spontaneous or mediated through cardiac pacing), results in greater changes in QT interval and T-wave amplitude in CRDS. This effective clinical diagnostic test for CRDS has become an important part of the evaluation of cardiac arrest.38
ICD
The insertion of an ICD is the standard treatment for all idiopathic VF patients with an aborted cardiac arrest or documented VF.3 There is a choice to be made between a subcutaneous or transvenous ICD. In a study by Kuschyk et al. of 183 patients, a majority of whom had either cardiac channelopathies or idiopathic VF, the authors showed that adverse defibrillator events, defined as a composite of inappropriate shocks and lead failure requiring surgical revision, were significantly lower in the subcutaneous ICD group compared to the transvenous ICD group (OR 0.55; 95% CI [0.41–0.72]) during a mean follow-up of 4.3 years.39 Programming should aim at avoiding inappropriate shocks with a single therapy zone at a high cutoff rate and long detection intervals. Recurrence of idiopathic VF with appropriate ICD therapies occurs in 18–39% of patients over a median follow-up of 41–63 months, with a median delay to the first recurrence of 4–12 months.13,18
Drug Therapy
Quinidine is indicated in cases of frequent VF recurrence or refusal of ICD implantation. It is a Class IA (INa-blocker) antiarrhythmic drug that also inhibits Ito. In some cases, quinidine is highly efficacious in preventing recurrences. The mean effective dose is 1,500 mg/day, corresponding to a serum quinidine level of 3.4 ± 1.6 mg/l. However, quinidine is not available worldwide.40
When facing an electrical storm defined by ≥3 VF episodes in 24 hours, oral quinidine and/or continuous infusion of isoproterenol can be effective. Electrical storms occur in up to 13% of patients with idiopathic VF. Isoproterenol should be initiated at 1–5 µg/min and titrated according to the clinical response (100–120/min sinus rate target). In a review by Belhassen et al., isoproterenol only had a 36% success rate in arrhythmia control.25
β-blockers, calcium inhibitors, and Class III drugs are usually ineffective in idiopathic VF.
Ablation Strategy
VF ablation is performed in a limited number of centres across the world. It should be proposed to a patient presenting VF recurrences or an electrical storm. It may also be proposed as a first-line treatment, in addition to an ICD, as the recurrence rate in idiopathic VF is high. The main challenge in these procedures is the identification of an adequate ablation target. In all cases, ablation should be extended approximately 1–2 cm² around the target site.41
Ablation of Purkinje Tissue
Purkinje arrhythmias are characterised by random occurrences, which complexifies the catheter ablation procedure. We recently described our approach for the ablation of these patients.41 The ideal configuration is to have PVC documentation and demonstration of causality. Prolonged 12-lead monitoring is of paramount importance. However, spontaneous and isolated Purkinje ectopy occurring after the VF episode are not necessarily causal. During a VF storm or shortly after a VF episode, non-sustained VF may be recorded, and the triggering PVC can be considered as the ablation target (Figure 3).
In the absence of PVC documentation, one should consider the usual suspects: moderator band, false tendons and papillary muscles, where Purkinje prepotentials can be identified. Pharmacological challenges, notably ajmaline, may elicit short-coupled PVCs in patients with idiopathic VF. Catecholamine tests have a lower inducibility rate. Isoprenaline induces Purkinje ectopy in 22% of cases, especially in the subgroup of patients presenting PVCs under sympathetic stress.
Intracardiac mapping should strive to identify Purkinje potentials to obtain a complete map of Purkinje tissue. Special care should be taken to avoid inadvertent bumping of the right and left bundles. The distal Purkinje tissue, characterised by a sharp signal with Purkinje-ventricle intervals ≤10 ms in sinus rhythm, can be targeted. During PVCs, Purkinje potentials precede the local electrogram by variable intervals that are usually ≥15 ms. Purkinje signals may present a diseased aspect with multiple fragmented components. When observed, these diseased signals should be targeted by ablation. The occurrence of QRS widening during ablation indicates potential catheter displacement toward the more proximal conduction system and ablation should be stopped. These intraventricular conduction disorders are sometimes difficult to avoid, and we reported the occurrence of transient left bundle branch block in one patient and nonspecific intraventricular conduction defects in six of 38 patients.
Pace-mapping is a useful tool to determine the ablation target as Purkinje ectopy is infrequent. However, the morphology of the Purkinje-triggered PVCs is rarely perfectly reproduced due to simultaneous capture of surrounding myocardium. Thus, pace mapping is ideally performed with a low pacing output (twice the diastolic threshold).
The ablation endpoint is often elusive. PVC suppression is only possible when facing spontaneous PVCs at the beginning of the procedure. The recurrence of PVCs does not necessarily translate into VF as the ablation of surrounding Purkinje fibres can suffice to impede initiation of VF. The ablation of key parts of the arborisation surrounding the PVC focus may minimise or suppress re-entry leading to repetitive beats. Nogami et al. also described the case of a patient with left-sided Purkinje ectopy in whom VF suppression was achieved with catheter ablation of the Purkinje’s network at a site distant from earliest activation.42
Intracardiac echocardiography appears mandatory for mapping and ablation. Indeed, string-like structures such as false tendons or the moderator band are rarely visible on standard electro-anatomic maps.
Microstructural Substrate Ablation
A detailed endocardial and epicardial mapping may reveal microstructural substrate alterations. In a recent report, we showed that areas of altered myocardial conduction are present in up to two-thirds of idiopathic VF patients and prove to be an efficient target for ablation.
The Near Future
Idiopathic VF is an unmet clinical challenge. Despite comprehensive attempts at clinical phenotype–genotype relationships, 50% of patients with primary electrical sudden cardiac arrest do not meet diagnostic criteria for other arrhythmic syndromes and are labelled with a diagnosis of idiopathic VF. Two axes of development are already in progress and should impact the understanding and management of idiopathic VF in the future.
Pathophysiology
Cellular Models and Mechanisms
Recently, Reilly et al. used idiopathic VF patient-specific induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) and optical mapping to characterise the cellular phenotype. They showed that idiopathic VF iPSC-CMs have abnormal calcium handling and increased proarrhythmic activity.43
iPSC-CMs enable patient-specific functional studies and would help to unveil phenotype-genotype relations. Omics approaches could then be used to broaden our molecular understanding of idiopathic VF.
Tissular Models and Mechanisms
Recent work has shown the implication of intracavitary structures, such as the moderator band, papillary muscles and false tendons, in several idiopathic VF case reports.44–49 Their precise role in arrhythmogenesis is not completely understood, and so further studies are needed. In vivo animal models are useful for this task, to study the underlying mechanisms associated with these structures.50 In complement, in silico modelling may help to understand the interplay between these structures and the initiation and maintenance of VF.
Technological Developments
Pulsed-field Ablation
Studies evaluating the impact of electroporation on Purkinje tissue are limited and provide contradictory results. Koruth et al. showed selective sparing of Purkinje’s fibres with pulsed-field myocardial ablation in a swine model.51 The authors performed 26 lesions using a multi-electrode catheter (Faraflex, Farapulse) in swine ventricles. Histology after 4 weeks revealed a single image of viable Purkinje fibres, despite the ablation of adjacent cardiomyocytes. This suggests that Purkinje fibres may have a lower susceptibility than cardiomyocytes to pulsed-field ablation.
These results contradict previous data by Livia et al., who performed pulsed-field ablation in eight canine hearts. They found that Purkinje tissue can be ablated with electroporation, and potentially has a lower threshold for pulsed-field ablation than ventricular myocytes.52 This apparent contradiction could possibly be explained by different ablation configurations.
Photon-counting CT
Photon-counting CT is a new technology enabling a better CT spatial resolution.53 It may allow the distinction of a specific ventricular anatomy related to idiopathic VF. Studies are needed to tackle this question.
Clinical Perspective
- Arrhythmogenic cardiomyopathy and electrical diseases represent a significant proportion of diagnoses attributed during follow-up in patients with an initial diagnosis of idiopathic VF.
- Genetic testing should be performed systematically.
- In addition to an ICD, catheter ablation should be proposed to a patient presenting VF recurrences or an electrical storm, or as a first-line treatment.
- The main challenge is the identification of the ablation target.
- The usual suspects are the moderator band, false tendons and papillary muscles.