- Email: [email protected]
Noninvasive clues for diagnosing ventricular tachycardia mechanism
Journal of Electrocardiology 51 (2018) 163–169
Contents lists available at ScienceDirect
Journal of Electrocardiology journal homepage: www.jecgonline.com
Review
Noninvasive clues for diagnosing ventricular tachycardia mechanism☆,☆☆ Andres Enriquez, MD, Michael Riley, MD, Francis Marchlinski, MD ⁎ Section of Cardiac Electrophysiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
a r t i c l e
i n f o
Keywords: Ventricular tachycardia Abnormal automaticity Triggered activity Reentry Electrocardiography
a b s t r a c t The electrophysiologic mechanisms responsible for the initiation and maintenance of ventricular tachycardia (VT) include enhanced automaticity, triggered activity and reentry. Differentiating between these three mechanisms can be challenging for the clinician and usually requires an invasive electrophysiology study. Establishing the underlying VT mechanism in a particular patient is helpful to define the optimal therapeutic approach, including the selection of pharmacologic agents or delineation of an ablation strategy. The purpose of this review is to provide insight into the possible VT mechanisms based on noninvasive clues from the clinical history, 12-lead electrocardiogram, tachycardia onset and termination and the response to pharmacologic manipulation. © 2017 Elsevier Inc. All rights reserved.
Introduction The electrophysiologic mechanisms responsible for ventricular tachycardia (VT) fall in one of 3 categories: [1] abnormal automaticity; [2] triggered activity; and [3] reentry [1]. (See Tables 1 and 2.) Understanding the cellular mechanism of VT in a particular patient is relevant for the prognosis, pharmacological management and also to define the optimal mapping and ablation strategies. Most focal VTs, due to a triggered or automatic mechanism, are amenable to betablockade therapy. In contrast, reentrant VTs usually require membrane active antiarrhythmic agents that slow conduction or prolong refractoriness to prevent reentry. In focal VTs the 12-lead electrocardiogram (ECG) provides a powerful tool for localizing the focal origin and the target for ablative therapy. In reentrant VTs the 12-lead ECG may only identify an exit for a larger macro-reentrant circuit. Activation mapping to define the site of earliest ventricular activation is the mainstay for mapping of focal VTs, typically complemented with pacemapping. On the other side, macro-reentrant VTs are better mapped with a combination of voltage mapping to delineate the arrhythmogenic substrate and entrainment maneuvers to define the critical components of the circuit. Focal VTs can be ablated with a single ablation lesion, whereas macro-reentrant VTs require linear lesions
☆ Funded in part by The Richard T. and Angela Clark Innovation Fund in Cardiac Electrophysiology. ☆☆ Disclosures: None. ⁎ Corresponding author at: 9 Founders Pavilion, Hospital of University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104, United States. E-mail address: [email protected] (F. Marchlinski).
https://doi.org/10.1016/j.jelectrocard.2017.11.009 0022-0736/© 2017 Elsevier Inc. All rights reserved.
between unexcitable boundaries aimed at interrupting the width of the reentrant circuit. Correct identification of the mechanism can be difficult in clinical practice and the 12-lead ECG by itself is limited for this purpose [2]. In this article we review some noninvasive clues from the clinical history, 12-lead ECG, telemetry monitoring or response to pharmacologic agents that may help the clinician to recognize the potential VT mechanism.
Pathophysiology Abnormal automaticity Automaticity is the property of cardiac cells to generate spontaneous action potentials and is the result of diastolic depolarization caused by a net inward current during phase 4 of the action potential [3]. Normal automaticity is a property of the sinoatrial and atrioventricular nodes and depends mainly on 2 phenomena: [1] diastolic activation of If (funny current), a mixed Na-K inward current, which unlike most voltage-sensitive currents, is activated by hyperpolarization rather than depolarization; and [2] release of calcium from the sarcoplasmic reticulum into the cytosol. The calcium, in turn, activates the Na+-Ca2+ exchanger, resulting in a net influx of sodium ions [4]. Ventricular myocardial cells do not display spontaneous diastolic depolarization or automaticity under normal conditions, but abnormal automaticity may occur under pathological conditions when the resting membrane potential becomes less negative. This may be consequence of a decrease in IK1 or an enhanced calcium release from the sarcoplasmic reticulum [4,5]. Similar to normal automaticity, abnormal automaticity is enhanced by β-adrenergic agonists and by reduction of external potassium.
164
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
Table 1 Electrophysiologic maneuvers for diagnosis of VT mechanism.
Initiation with PES Termination with PES Reset with fusion Entrainment with fusion Response to overdrive pacing Response to ventricular extrastimuli
Automaticity
Triggered activity
Reentry
No No No No Suppression Resetting with flat response
Sometimes Sometimes No No Acceleration/termination Resetting with flat/decreasing response
Yes Yes Yes Yes Entrainment/termination Resetting with flat/increasing/mixed response
Examples of abnormal automaticity include accelerated idioventricular rhythm in the setting of acute ischemia, myocarditis or cocaine intoxication [6,7]. Triggered activity It refers to action potential formation resulting from oscillations in membrane potential that are dependent on the preceding action potential (Fig. 1). When the amplitude of one of these afterdepolarizations reaches certain threshold, voltage-gated ion channels are activated, generating an action potential. Triggered activity can occur in the form of early or delayed afterdepolarizations. Early afterdepolarizations (EADs) occur during phase 2 or 3 of the cardiac action potential and are more manifest at slower heart rates, whereas delayed afterdepolarizations (DADs) occur during phase 4 of the action potential, after full repolarization, and are more dependent on faster heart rates. Examples of EADs include drug and electrolyte-induced Torsades de Pointes and some forms of polymorphic VT due to congenital long QT syndrome [8,9]. DADs are responsible for the majority of outflow tract VTs, catecholaminergic polymorphic ventricular tachycardia (CPVT) and ventricular arrhythmias associated with digitalis toxicity [10–14]. Reentry It is the most common mechanism of VT. It involves continuous repetitive propagation of an impulse around an area of anatomical or functional conduction block (Fig. 2). The following 3 criteria were originally proposed by Mines for identification of reentry: 1) unidirectional block must occur; 2) a region of slow conduction with return of the excitatory wave to its point of origin; and 3) interruption of the reentrant circuit at any points should terminate the tachycardia [15]. The substrate for reentry requires the presence of 2 pathways with different electrophysiologic properties separated by a central area of block (anatomical or functional). When an impulse encounters the central obstacle, unidirectional block occurs in one of the pathways and slow conduction occurs through the other pathway, creating a circus
Table 2 Noninvasive clues for diagnosis of VT mechanism.
Normal ECG Abnormal ECG (Q waves, epsilon waves, BBB) Outflow tract morphology Warm up/cool down Morphology of first beat Initiation during exercise N1 VT morphology Paired VT morphologies Adenosine
Catecholamine facilitation
Automaticity
Triggered activity
Reentry
++ 0
++ 0
+ +
0 + Identical to subsequent + − − Slowing or transient termination Increase
++ − Identical to subsequent ++ − − Termination
+ − Often different to subsequent + + + No effect
Increase
Increase/decrease
movement. For reentry to occur, the conduction within the unblocked pathway must be slow enough so the previously blocked pathway can recover its excitability by the time the reentrant wavefront returns. In other words, the anatomical length of the circuit should equal or exceed the reentrant wavelength. Reentrant arrhythmias can be reproducibly initiated and terminated by programmed stimulation. They can also interact with pacing and demonstrate the hallmark features of resetting and entrainment with fusion [16,17]. Resetting is the advancement of a tachycardia impulse by timed premature electrical stimuli. The extrastimulus is followed by an interval that is less than a fully compensatory pause before resumption of the original rhythm. Entrainment is the continuous resetting of a tachycardia circuit. During overdrive pacing the tachycardia is accelerated to the pacing rate, with resumption of the intrinsic rate upon abrupt cessation of pacing. Examples of reentry include: 1) scar-related VT in patients with structurally abnormal hearts due to ischemic or nonischemic cardiomyopathies; 2) bundle branch reentry, which is typically seen in patients with infrahisian conduction disease and involves antegrade conduction over the right bundle and retrograde conduction over the left bundle (or vice versa); 3) idiopathic left ventricular tachycardia (also known as fascicular VT, Belhassen VT or verapamil-sensitive VT), in which the macro-reentrant circuit involves the left posterior fascicle (or less commonly the left anterior fascicle) and abnormal slowly conducting Purkinje fibers; and 5) Phase 2 reentry associated with VAs in Brugada syndrome. Sinus rhythm ECG Baseline sinus rhythm 12-lead ECG may be helpful by indicating disease processes known to be associated with specific VT mechanisms. The presence of Q waves consistent with prior myocardial infarction indicates the substrate for scar-related reentry, especially if the VT morphology is consistent with an exit from the region of the infarct. Epsilon waves in the right precordial leads, especially in the setting of a left bundle-branch block VT, is a marker of arrhythmogenic right ventricular cardiomyopathy/dysplasia (ARVC/D) and suggests reentry involving non-ischemic scar localized to the right ventricle. Any evidence for His-Purkinje system disease as indexed by QRS widening, especially in the setting of a dilated cardiomyopathy, can predispose to bundle branch reentrant VT. A Brugada pattern is associated with phase 2 reentry, a special type of reentry proposed to be caused by heterogeneity in action potential distribution between the epicardium and endocardium [18]. A normal baseline ECG often reflects a structurally normal heart. Although VT due to any of the 3 mechanisms can occur in patients without structural heart disease, a normal baseline ECG coupled with specific morphologic patterns of VT may indicate a most likely mechanism: triggered activity for outflow tract tachycardias [12,13] and reentry involving the Purkinje system in fascicular VT [19]. Twelve-lead morphology of the ventricular tachycardia A 12-lead ECG recording of the VT allows to localize the site of origin or exit to a discrete region of the heart, however its value to define the VT mechanism is limited. Having said that, certain ECG patterns are
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
165
Fig. 1. Signal transduction pathways involved in cAMP-mediated triggered activity. Reproduced with permission from Springer, reference [1].
highly suggestive of a specific mechanism or can narrow among the mechanistic possibilities. Outflow tract VT and premature ventricular contractions (PVCs) characterized by large monophasic R waves in the inferior leads [20], especially in patients with structurally normal heart, can confidently be attributed to DAD-induced triggered activity. However, reentrant VTs associated with nonischemic cardiomyopathies can frequently originate near the perivalvular region [21]. As such, these tachycardias will mimic morphology of outflow tract VT due to a triggered mechanism. The presence of multiple VT morphologies and the identification of a region of low bipolar voltage surrounding these valvular structures support the diagnosis of nonischemic cardiomyopathy and a probable reentrant VT mechanism. A QRS morphology during VT typical for LBBB is suggestive of bundle branch reentrant VT. This typically occurs in the context of dilated cardiomyopathy with underlying His-Purkinje system disease [22]. The
VT is typically rapid and often presents with presyncope, syncope or cardiac arrest. A situation commonly encountered in patients with structural heart disease is the presence of more than one VT morphology. Paired VTs with similar cycle length but opposite axis, especially in the context of an ischemic etiology, points to macro-reentry involving a large circuit (Fig. 3). For example, 2 different VTs with LBBB pattern and left axis, one of them with a basal exit and another with apical axis, suggests macro-reentry involving a septal substrate. The presence of multiple LBBB VT morphologies with late precordial transition should suggest the possibility of ARVC. Most of the times reentrants VTs have stable cycle lengths, with little beat-to-beat variability. However, some cycle length variability can be observed initially during reentrant VTs and stability during VT is not a reliable feature to predict the mechanism. In other words, regularity does not equal reentry and irregularity does not equal triggered activity or automaticity.
Fig. 2. Reentrant circuit.
166
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
A particular and rare ECG pattern is bidirectional VT, defined as a tachycardia showing beat-to-beat alternation in the QRS axis. The most common causes of bidirectional VT are digitalis toxicity and CPVT, and the proposed mechanism is DAD-mediated triggered activity in anatomically separate parts of the conduction system [23]. Less likely, it may represent reentry around a circuit with two alternating exit sites. Tachycardia onset and termination The initiation and/or termination of VT can have important implications for mechanism (Figs. 4 and 5). Two different patterns of VT initiation have been described: [1] VT preceded by ventricular ectopy (single or multiple) of different morphology than that of the tachycardia; and [2] VT not preceded by ventricular ectopy (sudden onset) [24]. This information can be obtained from analysis of Holter recordings or stored electrograms of implantable cardiac devices. As previously mentioned, a hallmark of reentrant VTs is reproducible initiation with ventricular programmed stimulation. This type of initiation can also be seen in triggered VTs, but not in automatic VTs. Spontaneous PVCs are a noninvasive correlate of ventricular extrastimuli and, thus, initiation with a PVC of different morphology than that of the tachycardia suggests a reentrant mechanism. This in contrast with automatic and triggered VTs, which typically start with a beat similar to the ensuing beats of tachycardia. However, it must be said that sudden onset does not exclude reentry. This because a sinus impulse may conduct slowly through an area of diseased ventricular tissue and if there is unidirectional block in an adjacent limb of the tissue, reentry may occur as well. The first beat exiting the circuit is then the first beat of the tachycardia and would be expected to be late-coupled and of the same morphology as that of the VT [25]. Roelke et al. studied 73 ICD electrograms in 22 post-infarction patients with spontaneous
monomorphic VT and showed that morphology of the starting PVC was different to that of the ensuing tachycardia in 38 episodes (52%) and similar in 35 episodes (48%) [25]. In summary, consistent initiation with a PVC morphologically distinct to the VT typically will indicate reentry. On the other side, sudden initiation can be seen with any VT mechanism. The prototype of automatic VT is accelerated idioventricular tachycardia (AIVT), which is most often observed in the setting of acute myocardial infarction and reperfusion, but it can also be seen in acute myocarditis, hypertensive heart disease, digitalis intoxication and cocaine intoxication. It often begins as a late-coupled ventricular beat at a rate just faster than the preceding sinus rate. If either the sinus rate increases beyond the AIVT rate, or the AIVT rate slows, the VT is suppressed, only to reappear if either the sinus rate slows or the AIVT rate increases (Fig. 6, Panel A) [26]. A progressive increase in rate with tachycardia onset (warm-up) and/or progressive deceleration before tachycardia termination (cooldown) are also suggestive of an automatic mechanism (Fig. 6, Panel B). Finally, although less specific, the circumstances of tachycardia initiation may favor a particular mechanism. Triggered VTs are induced in the laboratory with rapid atrial or ventricular pacing and they are facilitated by cathecholamines, which increase the intracellular concentration of cyclic AMP. They typically occur during exercise or in the context of emotional stress, and are preceded by increasing heart rates. Reentrant arrhythmias, on the other side, are relatively independent of the autonomic tone or level of activity. Response to pharmacologic agents Probably the most useful and specific response to pharmacological intervention is the effect of adenosine on VT due to triggered activity. The cellular basis for triggered activity due to DAD is intracellular
Fig. 3. Paired VT morphologies in a patient with ischemic cardiomyopathy. The superior panel shows a right bundle left superior (RBLS) axis VT with a CL of 518 ms and the inferior panel shows a right bundle right inferior (RBRI) axis VT with a CL of 510 ms. This is manifestation of a large macro-reentrant circuit with exit sites mapped to the septal and lateral aspects of the ischemic scar.
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
167
Fig. 4. Examples of VT initiation with a PVC of different morphology than the ensuing beats of tachycardia (red arrow).
Fig. 5. Two examples of exercise-triggered VT mediated by DADs. Panel A shows initiation and termination of RVOT VT in a patient with structurally normal heart. Panel B shows CPVT in a child with a previous misdiagnosis of epilepsy. Note the classic sequence of sinus tachycardia, bidirectional VT and VF.
168
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
Fig. 6. Two examples of automatic ventricular arrhythmias. Panel A corresponds to a young patient without structural heart disease and shows an accelerated idioventricular rhythm which is suppressed when the sinus rate accelerates and reappears when the sinus rate slows down; note the presence fused beats (F) (courtesy Dr. Adrian Baranchuk). Panel B shows an automatic nonsustained VT in a patient with ischemic heart disease and prior myocardial infarction; note progressive acceleration and then deceleration before termination.
calcium overload, which induces a Ca2+-dependent depolarizing current (transient inward current or Iti), mainly given by activation of the Na+-Ca2+ exchanger [27]. The transient inward sodium current gives rise to DADs, which, if of sufficient amplitude, triggers a new action potential that may result in tachycardia. Catecholamines, through stimulation of the β-adrenergic receptor, causes activation of adenylyl cyclase (AC), increase of cAMP, activation of protein kinase A and phosphorylation of the L-type Ca channel, ryanodine receptor, and phospholamban. This results in increased intracellular calcium levels and enhanced activity of the Na+-Ca2+ exchanger. Adenosine exerts an inhibitory effect on AC and cAMP, reversing intracellular calcium overload. Given that adenosine has no antiarrhythmic effect in reentrant VT and only transiently suppresses (but does not terminate) automatic VT, termination of VT in response to adenosine is considered diagnostic of cAMP-mediated triggered activity [28]. In a recent study, Lerman et al. demonstrated that adenosine failed to terminate VT in 31 of 31 patients with reentrant VT due to structural heart disease, without effect on the VT cycle length [29]. In contrast, adenosine terminated VT in 45 of 50 (90%) patients with focal right or left outflow tract tachycardia. The sensitivity of adenosine for identifying triggered VT was 90% and its specificity was 100%. Another element to consider is the phenomenon of catecholamine facilitation of VT. In the same study 4 out 31 patients (13%) with reentrant VT due to structural heart disease showed catecholamine facilitation (3 by induction and 1 by perpetuation), whereas 78% of patients with outflow tract VT had catecholamine facilitation, all mediated via induction [29]. The effects of other drugs on VT are less specific to any one mechanism. Verapamil affects both outflow tract tachycardias due to triggered activity as well idiopathic forms of fascicular VT due to reentry. Class I and III antiarrhythmic drugs and betablockers also do not have specific effects targeting one or another mechanism.
Conclusion Identifying the electrophysiologic mechanism of VT is important to define a therapeutic strategy. Although this is challenging without an invasive electrophysiology study, several elements from the clinical history, 12-lead ECG in sinus rhythm or during tachycardia and analysis of Holter recordings and/or device electrograms, may provide insight about the most likely mechanism. Adenosine, due to its mechanismspecific effect is a valuable tool for the diagnosis of VT due to triggered activity. In our opinion, future lines of research should focus on new invasive and noninvasive technologies to understand the mechanism of VT in different cardiac conditions. In recent years, advances in electroanatomic mapping, cardiac imaging and the development of computer simulation models of ventricular arrhythmias have shown promising results in this regard.
References [1] Enriquez A, Frankel DS, Baranchuk A. Pathophysiology of ventricular tachyarrhythmias: from automaticity to reentry. Herzschrittmacherther Elektrophysiol 2017;28:149–56. [2] Riley MP, Marchlinski FE. ECG clues for diagnosing ventricular tachycardia mechanism. J Cardiovasc Electrophysiol 2008;19:224–9. [3] Antzelevitch C, Burashnikov A. Overview of basic mechanisms of cardiac arrhythmia. Card Electrophysiol Clin 2011;3:23–45. [4] Li R, Chiamvimonvat N. Adrenergic signaling and cardiac ion channels. In: Zipes D, Jalife J, editors. Cardiac electrophysiology. from cell to bedside. 5th ed. Philadelphia: Saunders Elsevier; 2009. p. 373–80. [5] Hai J, Albertson T, Evjen J, Asirvatham S. Mechanisms of ventricular tachycardia: impact for successful mapping and ablation. In: Mahapatra S, Marchlinski F, Natale A, Shivkumar K, editors. VT ablation: a practical guide. Minneapolis: Cardiotext; 2014. p. 25–39. [6] Grimm W. Accelerated idioventricular rhythm. Card Electrophysiol Rev 2001;5:328–31. [7] Tai YT, Lau CP, Fong PC, Li JP, Lee KL. Incessant automatic ventricular tachycardia complicating acute coxsackie B myocarditis. Cardiology 1992;80:339–44.
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169 [8] Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med 2004;350: 1013–22. [9] El-Sherif N. Mechanism of ventricular arrhythmias in the long QT syndrome: on hermeneutics. J Cardiovasc Electrophysiol 2001;12:973–6. [10] Rosen MR, Gelband H, Merker C, Hoffman BF. Mechanisms of digitalis toxicity. Effects of ouabain on phase four of canine Purkinje fiber transmembrane potentials. Circulation 1973;47:681–9. [11] Wieland JM, Marchlinski FE. Electrocardiographic response of digoxin-toxic fascicular tachycardia to Fab fragments: implications for tachycardia mechanism. Pacing Clin Electrophysiol 1986;9:727–38. [12] Lerman BB, Stein K, Engelstein ED, Battleman DS, Lippman N, Bei D, et al. Mechanism of repetitive monomorphic ventricular tachycardia. Circulation 1995;92:421–9. [13] Iwai S, Cantillon DJ, Kim RJ, Markowitz SM, Mittal S, Stein KM, et al. Right and left ventricular outflow tract tachycardias: evidence for a common electrophysiologic mechanism. J Cardiovasc Electrophysiol 2006;17:1052–8. [14] Priori SG, Napolitano C, Tiso N, Memmi M, Vignati G, Bloise R, et al. Mutations in the cardiac ryanodine receptor gene (hRyR2) underlie catecholaminergic polymorphic ventricular tachycardia. Circulation 2001;103:196–200. [15] Mines GR. On dynamic equilibrium in the heart. J Physiol 1913;46:350–83. [16] Issa ZF, Miller JM, Zipes DP. Electrophysiological mechanisms of cardiac arrhythmias. Clinical arrhythmology and electrophysiology, a companion to Brawnwald's heart disease. 2nd ed. Philadelphia: Saunders; 2012. p. 36–61. [17] Gaztañaga L, Marchlinski FE, Betensky BP. Mechanisms of cardiac arrhythmias. Rev Esp Cardiol (Engl Ed) 2012;65:174–85. [18] Antzelevitch C. The Brugada syndrome: ionic basis and arrhythmia mechanisms. J Cardiovasc Electrophysiol 2001;12:268–72.
169
[19] Lin D, Hsia HH, Gerstenfeld EP, Dixit S, Callans DJ, Nayak H, et al. Idiopathic fascicular left ventricular tachycardia: linear ablation lesion strategy for noninducible or nonsustained tachycardia. Heart Rhythm 2005;2:934–9. [20] Dixit S, Gerstenfeld EP, Callans DJ, Marchlinski FE. Electrocardiographic patterns of superior right ventricular outflow tract tachycardias: distinguishing septal and free-wall sites of origin. J Cardiovasc Electrophysiol 2003;14:1–7. [21] Hsia HH, Callans DJ, Marchlinski FE. Characterization of endocardial electrophysiological substrate in patients with nonischemic cardiomyopathy and monomorphic ventricular tachycardia. Circulation 2003;108:704–10. [22] Balasundaram R, Rao HB, Kalavakolanu S, Narasimhan C. Catheter ablation of bundle branch reentrant ventricular tachycardia. Heart Rhythm 2008;5(6 Suppl):S68–72. [23] Baher AA, Uy M, Xie F, Garfinkel A, Qu Z, Weiss JN. Bidirectional ventricular tachycardia: ping pong in the His-Purkinje system. Heart Rhythm 2011;8:599–605. [24] Ulus T, Kudaiberdieva G, Gorenek B. The onset mechanisms of ventricular tachycardia. Int J Cardiol 2013;167:619–23. [25] Roelke M, Garan H, McGovern BA, Ruskin JN. Analysis of the initiation of spontaneous monomorphic ventricular tachycardia by stored intracardiac electrograms. J Am Coll Cardiol 1994;23:117–22. [26] Grimm W, Marchlinski FE. Accelerated idioventricular rhythm and bidirectional ventricular tachycardia. In: Zipes DP, Jalife J, editors. Cardiac electrophysiology: from cell to bedside. Philadelphia: Elsevier, Inc.; 2004. p. 700–4. [27] Belardinelli L, Shryock JC, Song Y, Wang D, Srinivas M. Ionic basis of the electrophysiological actions of adenosine on cardiomyocytes. FASEB J 1995;9:359–65. [28] Lerman BB. Mechanism, diagnosis, and treatment of outflow tract tachycardia. Nat Rev Cardiol 2015;12:597–608. [29] Lerman BB, Ip JE, Shah BK, Thomas G, Liu CF, Ciaccio EJ, et al. Mechanism-specific effects of adenosine on ventricular tachycardia. J Cardiovasc Electrophysiol 2014;25:1350–8.
Contents lists available at ScienceDirect
Journal of Electrocardiology journal homepage: www.jecgonline.com
Review
Noninvasive clues for diagnosing ventricular tachycardia mechanism☆,☆☆ Andres Enriquez, MD, Michael Riley, MD, Francis Marchlinski, MD ⁎ Section of Cardiac Electrophysiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, United States
a r t i c l e
i n f o
Keywords: Ventricular tachycardia Abnormal automaticity Triggered activity Reentry Electrocardiography
a b s t r a c t The electrophysiologic mechanisms responsible for the initiation and maintenance of ventricular tachycardia (VT) include enhanced automaticity, triggered activity and reentry. Differentiating between these three mechanisms can be challenging for the clinician and usually requires an invasive electrophysiology study. Establishing the underlying VT mechanism in a particular patient is helpful to define the optimal therapeutic approach, including the selection of pharmacologic agents or delineation of an ablation strategy. The purpose of this review is to provide insight into the possible VT mechanisms based on noninvasive clues from the clinical history, 12-lead electrocardiogram, tachycardia onset and termination and the response to pharmacologic manipulation. © 2017 Elsevier Inc. All rights reserved.
Introduction The electrophysiologic mechanisms responsible for ventricular tachycardia (VT) fall in one of 3 categories: [1] abnormal automaticity; [2] triggered activity; and [3] reentry [1]. (See Tables 1 and 2.) Understanding the cellular mechanism of VT in a particular patient is relevant for the prognosis, pharmacological management and also to define the optimal mapping and ablation strategies. Most focal VTs, due to a triggered or automatic mechanism, are amenable to betablockade therapy. In contrast, reentrant VTs usually require membrane active antiarrhythmic agents that slow conduction or prolong refractoriness to prevent reentry. In focal VTs the 12-lead electrocardiogram (ECG) provides a powerful tool for localizing the focal origin and the target for ablative therapy. In reentrant VTs the 12-lead ECG may only identify an exit for a larger macro-reentrant circuit. Activation mapping to define the site of earliest ventricular activation is the mainstay for mapping of focal VTs, typically complemented with pacemapping. On the other side, macro-reentrant VTs are better mapped with a combination of voltage mapping to delineate the arrhythmogenic substrate and entrainment maneuvers to define the critical components of the circuit. Focal VTs can be ablated with a single ablation lesion, whereas macro-reentrant VTs require linear lesions
☆ Funded in part by The Richard T. and Angela Clark Innovation Fund in Cardiac Electrophysiology. ☆☆ Disclosures: None. ⁎ Corresponding author at: 9 Founders Pavilion, Hospital of University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104, United States. E-mail address: [email protected] (F. Marchlinski).
https://doi.org/10.1016/j.jelectrocard.2017.11.009 0022-0736/© 2017 Elsevier Inc. All rights reserved.
between unexcitable boundaries aimed at interrupting the width of the reentrant circuit. Correct identification of the mechanism can be difficult in clinical practice and the 12-lead ECG by itself is limited for this purpose [2]. In this article we review some noninvasive clues from the clinical history, 12-lead ECG, telemetry monitoring or response to pharmacologic agents that may help the clinician to recognize the potential VT mechanism.
Pathophysiology Abnormal automaticity Automaticity is the property of cardiac cells to generate spontaneous action potentials and is the result of diastolic depolarization caused by a net inward current during phase 4 of the action potential [3]. Normal automaticity is a property of the sinoatrial and atrioventricular nodes and depends mainly on 2 phenomena: [1] diastolic activation of If (funny current), a mixed Na-K inward current, which unlike most voltage-sensitive currents, is activated by hyperpolarization rather than depolarization; and [2] release of calcium from the sarcoplasmic reticulum into the cytosol. The calcium, in turn, activates the Na+-Ca2+ exchanger, resulting in a net influx of sodium ions [4]. Ventricular myocardial cells do not display spontaneous diastolic depolarization or automaticity under normal conditions, but abnormal automaticity may occur under pathological conditions when the resting membrane potential becomes less negative. This may be consequence of a decrease in IK1 or an enhanced calcium release from the sarcoplasmic reticulum [4,5]. Similar to normal automaticity, abnormal automaticity is enhanced by β-adrenergic agonists and by reduction of external potassium.
164
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
Table 1 Electrophysiologic maneuvers for diagnosis of VT mechanism.
Initiation with PES Termination with PES Reset with fusion Entrainment with fusion Response to overdrive pacing Response to ventricular extrastimuli
Automaticity
Triggered activity
Reentry
No No No No Suppression Resetting with flat response
Sometimes Sometimes No No Acceleration/termination Resetting with flat/decreasing response
Yes Yes Yes Yes Entrainment/termination Resetting with flat/increasing/mixed response
Examples of abnormal automaticity include accelerated idioventricular rhythm in the setting of acute ischemia, myocarditis or cocaine intoxication [6,7]. Triggered activity It refers to action potential formation resulting from oscillations in membrane potential that are dependent on the preceding action potential (Fig. 1). When the amplitude of one of these afterdepolarizations reaches certain threshold, voltage-gated ion channels are activated, generating an action potential. Triggered activity can occur in the form of early or delayed afterdepolarizations. Early afterdepolarizations (EADs) occur during phase 2 or 3 of the cardiac action potential and are more manifest at slower heart rates, whereas delayed afterdepolarizations (DADs) occur during phase 4 of the action potential, after full repolarization, and are more dependent on faster heart rates. Examples of EADs include drug and electrolyte-induced Torsades de Pointes and some forms of polymorphic VT due to congenital long QT syndrome [8,9]. DADs are responsible for the majority of outflow tract VTs, catecholaminergic polymorphic ventricular tachycardia (CPVT) and ventricular arrhythmias associated with digitalis toxicity [10–14]. Reentry It is the most common mechanism of VT. It involves continuous repetitive propagation of an impulse around an area of anatomical or functional conduction block (Fig. 2). The following 3 criteria were originally proposed by Mines for identification of reentry: 1) unidirectional block must occur; 2) a region of slow conduction with return of the excitatory wave to its point of origin; and 3) interruption of the reentrant circuit at any points should terminate the tachycardia [15]. The substrate for reentry requires the presence of 2 pathways with different electrophysiologic properties separated by a central area of block (anatomical or functional). When an impulse encounters the central obstacle, unidirectional block occurs in one of the pathways and slow conduction occurs through the other pathway, creating a circus
Table 2 Noninvasive clues for diagnosis of VT mechanism.
Normal ECG Abnormal ECG (Q waves, epsilon waves, BBB) Outflow tract morphology Warm up/cool down Morphology of first beat Initiation during exercise N1 VT morphology Paired VT morphologies Adenosine
Catecholamine facilitation
Automaticity
Triggered activity
Reentry
++ 0
++ 0
+ +
0 + Identical to subsequent + − − Slowing or transient termination Increase
++ − Identical to subsequent ++ − − Termination
+ − Often different to subsequent + + + No effect
Increase
Increase/decrease
movement. For reentry to occur, the conduction within the unblocked pathway must be slow enough so the previously blocked pathway can recover its excitability by the time the reentrant wavefront returns. In other words, the anatomical length of the circuit should equal or exceed the reentrant wavelength. Reentrant arrhythmias can be reproducibly initiated and terminated by programmed stimulation. They can also interact with pacing and demonstrate the hallmark features of resetting and entrainment with fusion [16,17]. Resetting is the advancement of a tachycardia impulse by timed premature electrical stimuli. The extrastimulus is followed by an interval that is less than a fully compensatory pause before resumption of the original rhythm. Entrainment is the continuous resetting of a tachycardia circuit. During overdrive pacing the tachycardia is accelerated to the pacing rate, with resumption of the intrinsic rate upon abrupt cessation of pacing. Examples of reentry include: 1) scar-related VT in patients with structurally abnormal hearts due to ischemic or nonischemic cardiomyopathies; 2) bundle branch reentry, which is typically seen in patients with infrahisian conduction disease and involves antegrade conduction over the right bundle and retrograde conduction over the left bundle (or vice versa); 3) idiopathic left ventricular tachycardia (also known as fascicular VT, Belhassen VT or verapamil-sensitive VT), in which the macro-reentrant circuit involves the left posterior fascicle (or less commonly the left anterior fascicle) and abnormal slowly conducting Purkinje fibers; and 5) Phase 2 reentry associated with VAs in Brugada syndrome. Sinus rhythm ECG Baseline sinus rhythm 12-lead ECG may be helpful by indicating disease processes known to be associated with specific VT mechanisms. The presence of Q waves consistent with prior myocardial infarction indicates the substrate for scar-related reentry, especially if the VT morphology is consistent with an exit from the region of the infarct. Epsilon waves in the right precordial leads, especially in the setting of a left bundle-branch block VT, is a marker of arrhythmogenic right ventricular cardiomyopathy/dysplasia (ARVC/D) and suggests reentry involving non-ischemic scar localized to the right ventricle. Any evidence for His-Purkinje system disease as indexed by QRS widening, especially in the setting of a dilated cardiomyopathy, can predispose to bundle branch reentrant VT. A Brugada pattern is associated with phase 2 reentry, a special type of reentry proposed to be caused by heterogeneity in action potential distribution between the epicardium and endocardium [18]. A normal baseline ECG often reflects a structurally normal heart. Although VT due to any of the 3 mechanisms can occur in patients without structural heart disease, a normal baseline ECG coupled with specific morphologic patterns of VT may indicate a most likely mechanism: triggered activity for outflow tract tachycardias [12,13] and reentry involving the Purkinje system in fascicular VT [19]. Twelve-lead morphology of the ventricular tachycardia A 12-lead ECG recording of the VT allows to localize the site of origin or exit to a discrete region of the heart, however its value to define the VT mechanism is limited. Having said that, certain ECG patterns are
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
165
Fig. 1. Signal transduction pathways involved in cAMP-mediated triggered activity. Reproduced with permission from Springer, reference [1].
highly suggestive of a specific mechanism or can narrow among the mechanistic possibilities. Outflow tract VT and premature ventricular contractions (PVCs) characterized by large monophasic R waves in the inferior leads [20], especially in patients with structurally normal heart, can confidently be attributed to DAD-induced triggered activity. However, reentrant VTs associated with nonischemic cardiomyopathies can frequently originate near the perivalvular region [21]. As such, these tachycardias will mimic morphology of outflow tract VT due to a triggered mechanism. The presence of multiple VT morphologies and the identification of a region of low bipolar voltage surrounding these valvular structures support the diagnosis of nonischemic cardiomyopathy and a probable reentrant VT mechanism. A QRS morphology during VT typical for LBBB is suggestive of bundle branch reentrant VT. This typically occurs in the context of dilated cardiomyopathy with underlying His-Purkinje system disease [22]. The
VT is typically rapid and often presents with presyncope, syncope or cardiac arrest. A situation commonly encountered in patients with structural heart disease is the presence of more than one VT morphology. Paired VTs with similar cycle length but opposite axis, especially in the context of an ischemic etiology, points to macro-reentry involving a large circuit (Fig. 3). For example, 2 different VTs with LBBB pattern and left axis, one of them with a basal exit and another with apical axis, suggests macro-reentry involving a septal substrate. The presence of multiple LBBB VT morphologies with late precordial transition should suggest the possibility of ARVC. Most of the times reentrants VTs have stable cycle lengths, with little beat-to-beat variability. However, some cycle length variability can be observed initially during reentrant VTs and stability during VT is not a reliable feature to predict the mechanism. In other words, regularity does not equal reentry and irregularity does not equal triggered activity or automaticity.
Fig. 2. Reentrant circuit.
166
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
A particular and rare ECG pattern is bidirectional VT, defined as a tachycardia showing beat-to-beat alternation in the QRS axis. The most common causes of bidirectional VT are digitalis toxicity and CPVT, and the proposed mechanism is DAD-mediated triggered activity in anatomically separate parts of the conduction system [23]. Less likely, it may represent reentry around a circuit with two alternating exit sites. Tachycardia onset and termination The initiation and/or termination of VT can have important implications for mechanism (Figs. 4 and 5). Two different patterns of VT initiation have been described: [1] VT preceded by ventricular ectopy (single or multiple) of different morphology than that of the tachycardia; and [2] VT not preceded by ventricular ectopy (sudden onset) [24]. This information can be obtained from analysis of Holter recordings or stored electrograms of implantable cardiac devices. As previously mentioned, a hallmark of reentrant VTs is reproducible initiation with ventricular programmed stimulation. This type of initiation can also be seen in triggered VTs, but not in automatic VTs. Spontaneous PVCs are a noninvasive correlate of ventricular extrastimuli and, thus, initiation with a PVC of different morphology than that of the tachycardia suggests a reentrant mechanism. This in contrast with automatic and triggered VTs, which typically start with a beat similar to the ensuing beats of tachycardia. However, it must be said that sudden onset does not exclude reentry. This because a sinus impulse may conduct slowly through an area of diseased ventricular tissue and if there is unidirectional block in an adjacent limb of the tissue, reentry may occur as well. The first beat exiting the circuit is then the first beat of the tachycardia and would be expected to be late-coupled and of the same morphology as that of the VT [25]. Roelke et al. studied 73 ICD electrograms in 22 post-infarction patients with spontaneous
monomorphic VT and showed that morphology of the starting PVC was different to that of the ensuing tachycardia in 38 episodes (52%) and similar in 35 episodes (48%) [25]. In summary, consistent initiation with a PVC morphologically distinct to the VT typically will indicate reentry. On the other side, sudden initiation can be seen with any VT mechanism. The prototype of automatic VT is accelerated idioventricular tachycardia (AIVT), which is most often observed in the setting of acute myocardial infarction and reperfusion, but it can also be seen in acute myocarditis, hypertensive heart disease, digitalis intoxication and cocaine intoxication. It often begins as a late-coupled ventricular beat at a rate just faster than the preceding sinus rate. If either the sinus rate increases beyond the AIVT rate, or the AIVT rate slows, the VT is suppressed, only to reappear if either the sinus rate slows or the AIVT rate increases (Fig. 6, Panel A) [26]. A progressive increase in rate with tachycardia onset (warm-up) and/or progressive deceleration before tachycardia termination (cooldown) are also suggestive of an automatic mechanism (Fig. 6, Panel B). Finally, although less specific, the circumstances of tachycardia initiation may favor a particular mechanism. Triggered VTs are induced in the laboratory with rapid atrial or ventricular pacing and they are facilitated by cathecholamines, which increase the intracellular concentration of cyclic AMP. They typically occur during exercise or in the context of emotional stress, and are preceded by increasing heart rates. Reentrant arrhythmias, on the other side, are relatively independent of the autonomic tone or level of activity. Response to pharmacologic agents Probably the most useful and specific response to pharmacological intervention is the effect of adenosine on VT due to triggered activity. The cellular basis for triggered activity due to DAD is intracellular
Fig. 3. Paired VT morphologies in a patient with ischemic cardiomyopathy. The superior panel shows a right bundle left superior (RBLS) axis VT with a CL of 518 ms and the inferior panel shows a right bundle right inferior (RBRI) axis VT with a CL of 510 ms. This is manifestation of a large macro-reentrant circuit with exit sites mapped to the septal and lateral aspects of the ischemic scar.
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
167
Fig. 4. Examples of VT initiation with a PVC of different morphology than the ensuing beats of tachycardia (red arrow).
Fig. 5. Two examples of exercise-triggered VT mediated by DADs. Panel A shows initiation and termination of RVOT VT in a patient with structurally normal heart. Panel B shows CPVT in a child with a previous misdiagnosis of epilepsy. Note the classic sequence of sinus tachycardia, bidirectional VT and VF.
168
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169
Fig. 6. Two examples of automatic ventricular arrhythmias. Panel A corresponds to a young patient without structural heart disease and shows an accelerated idioventricular rhythm which is suppressed when the sinus rate accelerates and reappears when the sinus rate slows down; note the presence fused beats (F) (courtesy Dr. Adrian Baranchuk). Panel B shows an automatic nonsustained VT in a patient with ischemic heart disease and prior myocardial infarction; note progressive acceleration and then deceleration before termination.
calcium overload, which induces a Ca2+-dependent depolarizing current (transient inward current or Iti), mainly given by activation of the Na+-Ca2+ exchanger [27]. The transient inward sodium current gives rise to DADs, which, if of sufficient amplitude, triggers a new action potential that may result in tachycardia. Catecholamines, through stimulation of the β-adrenergic receptor, causes activation of adenylyl cyclase (AC), increase of cAMP, activation of protein kinase A and phosphorylation of the L-type Ca channel, ryanodine receptor, and phospholamban. This results in increased intracellular calcium levels and enhanced activity of the Na+-Ca2+ exchanger. Adenosine exerts an inhibitory effect on AC and cAMP, reversing intracellular calcium overload. Given that adenosine has no antiarrhythmic effect in reentrant VT and only transiently suppresses (but does not terminate) automatic VT, termination of VT in response to adenosine is considered diagnostic of cAMP-mediated triggered activity [28]. In a recent study, Lerman et al. demonstrated that adenosine failed to terminate VT in 31 of 31 patients with reentrant VT due to structural heart disease, without effect on the VT cycle length [29]. In contrast, adenosine terminated VT in 45 of 50 (90%) patients with focal right or left outflow tract tachycardia. The sensitivity of adenosine for identifying triggered VT was 90% and its specificity was 100%. Another element to consider is the phenomenon of catecholamine facilitation of VT. In the same study 4 out 31 patients (13%) with reentrant VT due to structural heart disease showed catecholamine facilitation (3 by induction and 1 by perpetuation), whereas 78% of patients with outflow tract VT had catecholamine facilitation, all mediated via induction [29]. The effects of other drugs on VT are less specific to any one mechanism. Verapamil affects both outflow tract tachycardias due to triggered activity as well idiopathic forms of fascicular VT due to reentry. Class I and III antiarrhythmic drugs and betablockers also do not have specific effects targeting one or another mechanism.
Conclusion Identifying the electrophysiologic mechanism of VT is important to define a therapeutic strategy. Although this is challenging without an invasive electrophysiology study, several elements from the clinical history, 12-lead ECG in sinus rhythm or during tachycardia and analysis of Holter recordings and/or device electrograms, may provide insight about the most likely mechanism. Adenosine, due to its mechanismspecific effect is a valuable tool for the diagnosis of VT due to triggered activity. In our opinion, future lines of research should focus on new invasive and noninvasive technologies to understand the mechanism of VT in different cardiac conditions. In recent years, advances in electroanatomic mapping, cardiac imaging and the development of computer simulation models of ventricular arrhythmias have shown promising results in this regard.
References [1] Enriquez A, Frankel DS, Baranchuk A. Pathophysiology of ventricular tachyarrhythmias: from automaticity to reentry. Herzschrittmacherther Elektrophysiol 2017;28:149–56. [2] Riley MP, Marchlinski FE. ECG clues for diagnosing ventricular tachycardia mechanism. J Cardiovasc Electrophysiol 2008;19:224–9. [3] Antzelevitch C, Burashnikov A. Overview of basic mechanisms of cardiac arrhythmia. Card Electrophysiol Clin 2011;3:23–45. [4] Li R, Chiamvimonvat N. Adrenergic signaling and cardiac ion channels. In: Zipes D, Jalife J, editors. Cardiac electrophysiology. from cell to bedside. 5th ed. Philadelphia: Saunders Elsevier; 2009. p. 373–80. [5] Hai J, Albertson T, Evjen J, Asirvatham S. Mechanisms of ventricular tachycardia: impact for successful mapping and ablation. In: Mahapatra S, Marchlinski F, Natale A, Shivkumar K, editors. VT ablation: a practical guide. Minneapolis: Cardiotext; 2014. p. 25–39. [6] Grimm W. Accelerated idioventricular rhythm. Card Electrophysiol Rev 2001;5:328–31. [7] Tai YT, Lau CP, Fong PC, Li JP, Lee KL. Incessant automatic ventricular tachycardia complicating acute coxsackie B myocarditis. Cardiology 1992;80:339–44.
A. Enriquez et al. / Journal of Electrocardiology 51 (2018) 163–169 [8] Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med 2004;350: 1013–22. [9] El-Sherif N. Mechanism of ventricular arrhythmias in the long QT syndrome: on hermeneutics. J Cardiovasc Electrophysiol 2001;12:973–6. [10] Rosen MR, Gelband H, Merker C, Hoffman BF. Mechanisms of digitalis toxicity. Effects of ouabain on phase four of canine Purkinje fiber transmembrane potentials. Circulation 1973;47:681–9. [11] Wieland JM, Marchlinski FE. Electrocardiographic response of digoxin-toxic fascicular tachycardia to Fab fragments: implications for tachycardia mechanism. Pacing Clin Electrophysiol 1986;9:727–38. [12] Lerman BB, Stein K, Engelstein ED, Battleman DS, Lippman N, Bei D, et al. Mechanism of repetitive monomorphic ventricular tachycardia. Circulation 1995;92:421–9. [13] Iwai S, Cantillon DJ, Kim RJ, Markowitz SM, Mittal S, Stein KM, et al. Right and left ventricular outflow tract tachycardias: evidence for a common electrophysiologic mechanism. J Cardiovasc Electrophysiol 2006;17:1052–8. [14] Priori SG, Napolitano C, Tiso N, Memmi M, Vignati G, Bloise R, et al. Mutations in the cardiac ryanodine receptor gene (hRyR2) underlie catecholaminergic polymorphic ventricular tachycardia. Circulation 2001;103:196–200. [15] Mines GR. On dynamic equilibrium in the heart. J Physiol 1913;46:350–83. [16] Issa ZF, Miller JM, Zipes DP. Electrophysiological mechanisms of cardiac arrhythmias. Clinical arrhythmology and electrophysiology, a companion to Brawnwald's heart disease. 2nd ed. Philadelphia: Saunders; 2012. p. 36–61. [17] Gaztañaga L, Marchlinski FE, Betensky BP. Mechanisms of cardiac arrhythmias. Rev Esp Cardiol (Engl Ed) 2012;65:174–85. [18] Antzelevitch C. The Brugada syndrome: ionic basis and arrhythmia mechanisms. J Cardiovasc Electrophysiol 2001;12:268–72.
169
[19] Lin D, Hsia HH, Gerstenfeld EP, Dixit S, Callans DJ, Nayak H, et al. Idiopathic fascicular left ventricular tachycardia: linear ablation lesion strategy for noninducible or nonsustained tachycardia. Heart Rhythm 2005;2:934–9. [20] Dixit S, Gerstenfeld EP, Callans DJ, Marchlinski FE. Electrocardiographic patterns of superior right ventricular outflow tract tachycardias: distinguishing septal and free-wall sites of origin. J Cardiovasc Electrophysiol 2003;14:1–7. [21] Hsia HH, Callans DJ, Marchlinski FE. Characterization of endocardial electrophysiological substrate in patients with nonischemic cardiomyopathy and monomorphic ventricular tachycardia. Circulation 2003;108:704–10. [22] Balasundaram R, Rao HB, Kalavakolanu S, Narasimhan C. Catheter ablation of bundle branch reentrant ventricular tachycardia. Heart Rhythm 2008;5(6 Suppl):S68–72. [23] Baher AA, Uy M, Xie F, Garfinkel A, Qu Z, Weiss JN. Bidirectional ventricular tachycardia: ping pong in the His-Purkinje system. Heart Rhythm 2011;8:599–605. [24] Ulus T, Kudaiberdieva G, Gorenek B. The onset mechanisms of ventricular tachycardia. Int J Cardiol 2013;167:619–23. [25] Roelke M, Garan H, McGovern BA, Ruskin JN. Analysis of the initiation of spontaneous monomorphic ventricular tachycardia by stored intracardiac electrograms. J Am Coll Cardiol 1994;23:117–22. [26] Grimm W, Marchlinski FE. Accelerated idioventricular rhythm and bidirectional ventricular tachycardia. In: Zipes DP, Jalife J, editors. Cardiac electrophysiology: from cell to bedside. Philadelphia: Elsevier, Inc.; 2004. p. 700–4. [27] Belardinelli L, Shryock JC, Song Y, Wang D, Srinivas M. Ionic basis of the electrophysiological actions of adenosine on cardiomyocytes. FASEB J 1995;9:359–65. [28] Lerman BB. Mechanism, diagnosis, and treatment of outflow tract tachycardia. Nat Rev Cardiol 2015;12:597–608. [29] Lerman BB, Ip JE, Shah BK, Thomas G, Liu CF, Ciaccio EJ, et al. Mechanism-specific effects of adenosine on ventricular tachycardia. J Cardiovasc Electrophysiol 2014;25:1350–8.