Early repolarization patterns and the role of additional proarrhythmic triggers

Abstract

Early repolarization (ER) patterns have traditionally been regarded as benign variants and are in fact more likely to be observed in healthy athletes at times of peak fitness. In the vast majority of cases, ER is not a malignant variant per se, unless proven otherwise. However, recent case–control series and population studies have established a significant association between ER and an increased risk of arrhythmic death,^1,2 which has raised concern among physicians and arrhythmologists in particular.

Nevertheless, the estimation of the arrhythmic risk following the incidental discovery of notched or slurred J-waves in inferior or lateral leads and the distinction between benign and malignant electrocardiographic (ECG) variants of ER are still controversial subjects. In benign ER, the ST-segment and T-wave patterns have a relative temporal stability and T-waves are generally large, positive, and concordant relative to the ST-segment deviation from V2 to V4. Rapidly ascending ST-segment after the J-point, the dominant ST pattern in healthy athletes, seems to be a benign variant of ER as well.³ In contrast, in malignant ER syndromes, horizontal/descending ST variant is more prevalent, T-waves may be discordant to their respective ST-segment deviation, the ST-/T-waves are dynamic, inconstant, and dramatic changes in morphology frequently occur (predominantly at night during vagotonic predominance).^3,4 However, these features have not been unequivocally and consistently demonstrated. Rollin and colleagues proposed that a slurred ER pattern or ascendant ST-segment is not associated with increased mortality, whereas a notched J-wave and horizontal/descendant ST-segments were associated with the highest risk of all-cause and cardiovascular deaths. Early repolarization pattern localization, the amplitude of J-point elevation, and T-wave morphology did not help distinguish between benign and malignant forms or ER in their study. Data regarding arrhythmic mortality were not reported, which was a limitation of the study, considering the proposed mechanism of death in these patients [idiopathic ventricular fibrillation (VF)].² On the other hand, a previous study by Tikkanen et al.,⁵ including a larger cohort of patients and a longer follow-up, had suggested J-point elevation of >0.2 mV in inferior leads, but not the lateral leads, predicted death from cardiac or arrhythmic causes, while a slurred J-wave with J-point elevation >0.2 mV had a greater hazard ratio for arrhythmic death than the equivalent notched pattern. Despite these considerations, there is a common agreement that J-waves, particularly of large amplitude, dynamic, and present in multiple leads, are more prevalent in patients with idiopathic VF, but the distinction between benign and malignant variants is still an area of ongoing research.

Rosso et al.⁶ have made an effort into ‘putting the evidence into perspective’, suggesting the increased dispersion of repolarization in individuals with J-waves would place them at increased arrhythmic risk, but only in the presence of additional proarrhythmic factors or triggers.

The scientific community has recently embraced research for the identification of those proarrhythmic triggers that could unmask the malignant nature of certain ER variants. Acute or chronic myocardial ischaemia, the autonomic imbalance of chronic heart failure, and electrolyte disturbances such as severe hypokalaemia have claimed a role as such triggers.

Pei et al.⁷ investigated the relationship between particular ECG parameters and sudden cardiac death (SCD) in chronic heart failure and concluded that the presence of J-waves or fragmented QRS complexes in the inferior leads independently predicted a higher risk for SCD in ischaemic or non-ischaemic dilated cardiomyopathy. Autonomic imbalance, heterogeneous dispersion of repolarization, or fragmented late depolarization could be the mechanisms associated to such a finding. Hypokalaemia has also been shown to decrease the threshold for VF in the presence of ER.⁸ It is reported to produce a reduction in the epicardial but not the endocardial I_to in experimental studies and to prolong the action potential duration in Purkinje fibres, while shortening it in adjacent ventricular cells. The resulting increased dispersion of repolarization at the Purkinje–ventricular junction and amplified spatial repolarization gradients explain the increased risk for VF.⁹

Patel et al.¹⁰ suggested ER and, in particular, a notching morphology in the inferior leads to be associated with increased risk of life-threatening ventricular arrhythmias in patients with coronary artery disease, even after adjustment for left ventricular ejection fraction.

However, the prognostic role of ER in acute myocardial ischaemia has received much more emphasis than in any other clinical context. Tikkanen et al.^1,1 found a higher prevalence of ER in a standard 12-lead ECG in victims of SCD than in survivors of an acute coronary event, which suggested the presence of ER could increase the vulnerability to fatal arrhythmia during acute myocardial ischaemia. Such finding could provide a mechanistic link between this ECG pattern and higher arrhythmic mortality of middle-aged/elderly subjects. Rudic et al.¹² concluded that ER was more common in patients with myocardial infarction (MI) complicated by VF than in patients with MI without ventricular tachyarrhythmias, with no differences in the distribution of ER in the 12-lead ECG. A study by Naruse et al.¹³ suggested an increased risk of VF occurrence within 48 h of MI onset in patients with previous ER pattern. Patel et al.¹⁴ reached similar conclusions. Table 1 highlights the main findings of studies relating ER and SCD due to an acute coronary event.

The mechanistic basis for this apparent ER-related arrhythmogenesis is mostly unknown. In ER, the ST-segment elevation is attributed to the shortening of ventricular action potential duration in epicardial regions resulting in a current similar to the subepicardial injury currents of the Brugada syndrome (BrS) and idiopathic VF. In fact, the similarities between the BrS and ER syndrome in response to rate, neuromodulation, and pharmacological agents strongly suggest a parallelism of mechanisms. Predominance in young otherwise healthy males, predisposition to familial occurrence, and the dynamicity of the ECG findings are also common to both entities. Yet, discrete differences concerning lead localization of ECG abnormalities and opposite response to sodium channel blockers and body temperature (hypothermia accentuates J-point elevation, whereas hyperthermia often triggers the arrhythmic episodes of the BrS) raise some doubts regarding the extent of overlap between these two syndromes. Also, the much more pronounced transmural and epicardial dispersion of repolarization and refractoriness, setting the stage for both phase 2 re-entry and circus movement re-entry, explains the more severe arrhythmogenic nature of the BrS compared with the ER syndrome.¹⁵ Pérez-Riera et al. proposed the underlying mechanism for such a difference in clinical arrhythmogenesis among these two syndromes to be the difference in I_to density and I_to-mediated epicardial spike and dome configuration. When I_to is prominent, complete loss of the dome may occur due to either a decrease in inward currents or an increase in outward currents leading to phase 2 re-entry capable of initiating VF as in the BrS. When I_to is relatively small, as in the ER syndrome, partial depression of the dome occurs without the development of phase 2 re-entry.⁵ Therefore, under certain conditions known to predispose to ST-segment elevation, such as an ST-segment elevation MI, patients with an ER pattern could be at increased risk. Depression of the epicardial action potential plateau in ER creates a transmural gradient that may not be arrhythmogenic by itself. However, further increases in the net repolarizing current with subsequent loss of the epicardial action potential dome and a more profound dispersion of repolarization may lower the threshold for early after-depolarizations and create an optimal electric substrate for ventricular arrhythmogenesis.

The ‘repolarization disorder hypothesis’ could be weighed against the ‘depolarization disorder mechanism’, which has been substantiated by a number of clinical observations. Meregalli et al.¹⁶ suggested that the BrS was not fully explained by one single mechanism and therefore investigation should aim for clarification of the contribution of various pathophysiological mechanisms in BrS patients. General conduction slowing is a feature of BrS and supports the ‘depolarization hypothesis’. In fact, late potentials are highly prevalent in the BrS^17,18 and predict ventricular tachycardia (VT)/VF inducibility, as opposed to corrected QT dispersion and T-wave alternans.¹⁷ Furthermore, late potentials coincide with spontaneous ST elevation and late r' in V1–V3, which have been shown to correlate with VF onset.¹⁹ This conduction delay has been mapped to the right ventricular outflow tract (RVOT) and shown to increase with antiarrhythmic class I-C drug challenge and decrease with isoproterenol.²⁰ Echocardiographic studies using tissue Doppler have corroborated this finding, as the amplitude of ST elevation in BrS patients has been demonstrated to correlate with delay in right ventricular contraction.²¹ Catheter ablation over this abnormal area results in normalization of the Brugada ECG pattern and prevents inducible or spontaneous VT/VF episodes, as demonstrated by Nademanee et al.²² Based on the arrhythmogenic stimulus related to conduction delay in the RVOT and the occurrence of late potentials, some investigators have proposed depolarization abnormalities would be the electrogenic mechanism of the BrS, whereas ECG repolarization abnormalities (the coved-type ST-segment) would be the result of depolarization deviance in combination with mild structural abnormalities.²³ This topic remains a controversial subject and consensus has not been achieved.^16,23

Data on ER syndrome are lacking. Although the diverse similarities between the BrS and ER syndrome should not imply a substantial pathophysiological similarity, some authors have highlighted the importance of depolarization anomalies in idiopathic VF patients with J-waves. Abe et al. assessed the role of risk markers reflecting electrophysiological abnormalities in the pathophysiology of J-waves in patients with idiopathic VF. Repolarization abnormality markers such as T-wave alternans and QT dispersion did not differ between patients with and without J-waves, while the incidence of late potentials (a depolarization abnormality marker) in the idiopathic VF J-wave group was higher than in the non-J-wave group.²⁴ This theory does not seem to explain the potentially higher arrhythmic mortality of patients with ER patterns during acute myocardial ischaemia, as the presence of ventricular late potentials is not significantly associated with cardiac deaths or serious arrhythmic events in unselected post-infarction patients in the modern reperfusion era.²⁵ Even so, the contribution of various pathophysiological mechanisms in ER syndrome seems as likely as in the BrS.

Appearance or transient augmentation of global J-waves may be an important marker for lethal arrhythmias in acute ischaemia.²⁶ Ventricular tachycardia/VF initiation is more commonly associated with a short–long–short sequence, and premature ventricular complexes display a shorter coupling interval in patients with ER pattern compared with those with the BrS.²⁷ Potential documentation of the mode of onset of VF in patients with a MI and history of a ‘malignant’ ER pattern could add substantial data to this subject and reinforce the hypothesis that slurred or notched J-waves are directly and independently associated to an increased risk of sudden death in patients with an acute coronary event.

The potentially increased risk of acute ischaemia-related SCD in subjects with malignant ER patterns raises some questions regarding their optimal therapeutic management. Hopefully, future clinical research may address further issues such as the optimal target heart rate for secondary cardiovascular prevention in these patients or the potential indication for invasive risk stratification through electrophysiological study in patients with VF in the early phase of an acute MI. Complex ventricular arrhythmias may be a marker of a poorer long-term prognosis in patients with ER patterns even when occurring in the first hours of the MI index.

• More aggressive primary prevention of coronary artery disease and patient education about the importance of early hospital admission in case of symptoms suggestive of myocardial ischaemia may be warranted;

• Prevention and treatment of hypokalaemia and hypomagnesaemia may be as helpful as in patients presenting with polymorphic VT due to other pathological conditions.

• The anti-ischaemic benefit of lowering heart rate must be balanced against the risk of bradycardia-induced exacerbation of ER's arrhythmogenic potential. Notwithstanding this consideration, the undisputable evidence supporting the use of beta-blockers in patients with a MI must not be underrated.

• As the ventricular cycle length is often increased directly prior to polymorphic VTs, preventive pacing to avoid pauses and sudden rate accelerations may be considered in those patients with history of ER, a malignant ventricular arrhythmia at admission, and persistence of bradycardia, and/or accentuated notched, or slurred J-wave patterns.

Nevertheless, further studies are needed to clarify the role of ER in acute ischaemic syndromes. Caution should be raised not to ‘throw all eggs in one basket’, as asymptomatic subjects with isolated ER patterns are very unlikely to develop malignant ventricular arrhythmias or to suffer SCD. Undoubtedly, the majority of individuals with ER have a benign prognosis. Even so, it is instructive to remember that the BrS was regarded as benign for more than three decades and the ER ECG pattern can be readily converted to the Brugada one in experimental models.²⁸ Therefore, although caring physicians should be aware that there is no evidence at present to support different management for patients with ER patterns having an acute MI, clinical awareness of the potential arrhythmogenic role of ER or J-wave syndromes in this clinical context is warranted.

Judgment based on clinical presentation, history of syncope, or family history of SCD seems the most important risk stratification tool for patients with ER syndrome, but, although risk stratification of patients with ER and idiopathic VF is reasonably straightforward (survivors of cardiac arrest are potential candidates for implantable cardioverter defibrillator implantation), risk stratification is more complex and remains currently unproven in asymptomatic subjects.

Conflicts of interest: none declared.

References