Haris M.Haqqani, MBBS (Hons), PhD, and David J. Callans, MD
A 33-year-old construction worker with nonischemic dilated cardiomyopathy (DCM) presented with multiple implantable cardioverter-defibrillator (ICD) shocks due to recurrent monomorphic ventricular tachycardia (VT). He had mild global left ventricular (LV) systolic dysfunction with an ejection fraction (EF) of 40% and basal septal intramural delayed gadolinium enhancement (DGE) on magnetic resonance imaging (MRI). He had failed medical therapy with amiodarone and mexiletine and underwent catheter ablation. Three unmappable VT morphologies were induced, including a left bundle branch block (LBBB) morphology, VT of 325 ms cycle length that corresponded to the clinical VT on cycle length, and ICD electrogram analysis (VT 1). Electroanatomic voltage mapping was then performed to define the VT substrate with a 3.5-mm tip open-irrigated catheter using a point-by-point contact mapping technique (Figure 20-1). Endocardial bipolar voltage was normal (>1.5 mV) in both the left and right ventricles (RV) (Figure 20-1A).1 Epicardial bipolar voltage mapping showed normal voltage (>1.0 mV) in all areas except those overlying the coronary arteries in the interventricular and atrioventricular sulci (Figure 20-1B).2 This corresponds to the expected location of epicardial fat, which exerts an electrogram-attenuating effect on the signal generated by the underlying epimyocardium. No fractionated or isolated late potentials were seen to suggest the presence of epicardial fibrosis.2 Unipolar electrogram analysis of the RV endocardial map suggested a normal RV free-wall voltage (>5.5 mV) (Figure 20-1C).3Unipolar electrogram analysis of the LV did not suggest the presence of any basolateral LV intramural substrate; however, the basal septum exhibited low unipolar signal amplitude (<8.3 mV) from both the LV and RV aspect.4 This very likely corresponded to the septal area of DGE seen on the MRI (Figure 20-1D, arrows).4 The putative exit of all three VTs was mapped to the basal septal region using pace-mapping, and extensive substrate ablation here rendered the patient noninducible (Figure 20-2).
FIGURE 20-1 Electroanatomic voltage maps in a 33-year-old man with nonischemic idiopathic dilated cardiomyopathy.
FIGURE 20-2 Posteroanterior projection of unipolar biventricular voltage maps showing periannular septal and infundibular substrate. The three inducible VTs were pace-mapped to this region where radiofrequency ablation rendered them noninducible.
Catheter ablation of scar-related VT is often beset by the problem of multiple poorly tolerated, unstable VT morphologies that cannot be defined by entrainment mapping. These unmappable VTs need to be targeted in sinus rhythm, and their critical circuit components are invariably located in zones of fibrosis. The various forms of structural heart disease that cause scar-related VT usually display confluent areas of myocardial surface fibrosis. This involves the endocardium in ischemic (postinfarct) cardiomyopathy or the epicardium in arrhythmogenic right ventricular dysplasia. The various forms of nonischemic dilated cardiomyopathy (DCM) are characterized by diffuse interstitial and replacement fibrosis, and this may result in a basolateral, predominantly epicardial pattern scarring. However, the normal endocardial and epicardial bipolar surface voltage maps in a VT patient with DCM strongly suggested the presence of intramural substrate. Unipolar electrograms have a wider “field-of-view” and may be sensitive to deeper layers of fibrosis within the myocardium. In this case, substrate mapping disclosed the presence of unipolar low voltage on basal septum, suggestive of intramural scarring that corresponded to the DGE-defined fibrosis seen with MRI.
VENTRICULAR ELECTROGRAM RECORDING AND DEFINITIONS
Contact intracardiac electrograms are recorded from the ventricles by placing the distal recording electrodes of a mapping catheter at sequential endocardial or epicardial sites. Typical mapping catheters in current use employ a 3.5-mm distal tip electrode, a 2-mm ring electrode on the shaft, with a 1-mm interelectrode spacing. Unipolar electrograms can be recorded from each of these two electrodes using either Wilson’s central terminal or an indifferent electrode in the inferior vena cava. Bipolar electrograms are recorded between the tip and ring electrodes by vector summation of their individual unipolar signals. The signals are digitized and amplified and then subject to high and low pass filtration to remove the effects of low-frequency respiratory noise and high-frequency electrical noise from the final processed electrogram.
Cassidy et al found that normal ventricular bipolar electrograms recorded from healthy myocardial tissue (with a 10-mm bipole, filtered at 30-500 Hz) exhibit characteristic properties (Figure 20-3A). They have sharp, biphasic, or triphasic deflections with peak-to-peak amplitude ≥3 mV, duration <70 ms, and/or amplitude:duration ratio >0.045.5 As they are acquired, normal and abnormal electrograms can be plotted onto color-coded three-dimensional voltage maps using an electroanatomic mapping system.
FIGURE 20-3 Characteristics of normal, fractionated, isolated, and very late ventricular electrograms.
All forms of ventricular scarring cause disruption to the normal myocardial syncytial architecture, variably replacing necrotic or apoptotic cardiomyocytes with collagen. In dense sheets of scar, surviving myocyte bundles may be present with abnormally slow and circuitous electrical activation due to the insulating effects of confluent fibrotic zones as well as the loss of normal cell-to-cell coupling. This slow conduction can be also caused by more diffuse interstitial and replacement fibrosis. In either case, electrograms recorded from such regions display reduced peak-to-peak amplitude due to lesser action potential summation from the lower surviving myocyte mass. Additionally, the slow conduction caused by shards of fibrosis results in electrogram prolongation, with multiple deflections commonly seen due to adjacent myocyte bundles being sequentially activated in delayed fashion, rather than rapidly in parallel. These electrogram changes are known as fractionation, and fractioned ventricular electrograms display multiple deflections from baseline, an amplitude ≤0.5 mV, a duration ≥133 ms, and/or amplitude:duration ratio <0.005 (Figure 20-3B). If delayed conduction occurs into surviving myocyte bundles that are well insulted by surrounding collagen sheets, late components of the local electrogram may be inscribed at a discernible interval (with an isoelectric duration of >20 ms) after the initial electrogram component, which is usually a far-field signal (Figure 20-3C). In very slow conduction regions, these potentials may occur well after the surface QRS complex (Figure 20-3D). These are known variably as late potentials (LP) or isolated late potentials (ILP) and likely represent depolarization of anatomically constrained surviving fiber bundles within dense sheets of scar.6 Local abnormal ventricular activities (LAVA) is another term used to describe similar electrogram abnormalities within scar.7
BIPOLAR ENDOCARDIAL VOLTAGE MAPPING
Electrogram amplitude and morphology analysis and interpretation is greatly facilitated by the use of electroanatomic mapping systems. These systems all localize the tip of the mapping catheter in 3-dimensional space and can use this to create a virtual geometric shell of the ventricle. Electrogram information can be incorporated into this shell using color coding to generate 3-dimensional voltage maps (Figure 20-4).8 Using an electroanatomic mapping system and detailed point-by-point sinus rhythm mapping in patients without structural heart disease, Marchlinski et al were able to define the normal LV and RV endocardial bipolar voltage limits.1 They found that the mean bipolar electrogram amplitude in these normal ventricles was 4.8 ± 3.1 mV with 95% of LV recordings displaying a bipolar voltage >1.55 mV. Patients with prior myocardial infarct tend to have a relatively sharp demarcation between the normal ventricle and the area of scar. The border zone of the scar corresponds to the area of bipolar voltage ≤1.5 mV, but the majority of most postinfarct scars contain confluent areas of <0.5 mV, the so-called dense-scar region (Figure 20-4).1,9 Several chronic infarct large animal models have validated bipolar voltage mapping against pathologic analysis and imaging of the infarct scar. Callans et al found that the scar area as imaged using intracardiac echocardiography (ICE) corresponded to the area subtended by bipolar electrogram voltages of ≤2.0 mV, with the pathologically defined area correlating best with bipolar signal amplitude of ≤1.0 mV. Reddy et al were able to show that the area defined by bipolar electrogram amplitude of ≤1.5 mV correlated very well with the pathologically defined infarct (r = 0.96, P = 0.0007).10
FIGURE 20-4 Endocardial left ventricular substrate map of a 68-year-old man with severe ischemic cardiomyopathy (with three prior infarcts and bypass graft surgery), ejection fraction 20%, who presented in near incessant VT requiring multiple ICD shocks. He has a large area of anteroapicoseptal scar with multiple isolated late potentials (ILP) found throughout. In VT, the labeled ILP corresponded to a mid-diastolic potential (MDP) that was proven to be an entrance site to the common central isthmus. Ablation here terminated the tachycardia early, and further ablation rendered it noninducible.
EPICARDIAL VOLTAGE MAPPING
The advent of percutaneous epicardial mapping allowed for the successful ablation of VT associated with many nonischemic substrates (eg, Chagasic cardiomyopathy, idiopathic DCM, arrhythmogenic right ventricular dysplasia [ARVD]) in which epicardial VT circuit components are a common finding. While similar substrate mapping criteria could be used on the epicardium as on the endocardium, the presence of epicardial fat can act as a unique confounder by its potential insulating effect, leading to electrogram attenuation. Cano et al analyzed epicardial electrograms in 8 patients without structural heart disease and found a mean bipolar voltage of 3.2 ± 2.5 mV with 95% of all signals having a bipolar amplitude of >0.94 mV when signals over the atrioventricular (AV) grooves and large coronary vessels were excluded.2 On this basis, they defined a normal epicardial bipolar electrogram voltage to be >1.0 mV (Figure 20-5). More importantly, however, these investigators found that epicardial scar could not be defined on the basis of voltage alone and that electrogram analysis plays a pivotal role. In patients with idiopathic DCM, epicardial scar was strongly suggested by the presence of fractionated or isolated potentials, rather than just low voltage, as such signals were generally not found in the region of the AV grooves or the large coronary vessels (Figure 20-5).2
FIGURE 20-5 Epicardial voltage map of a 48-year-old man with dilated cardiomyopathy showing a periannular substrate distribution with isolated and fractionated electrograms recorded within basal low-voltage (bipolar voltage <1.0 mV) areas.
UNIPOLAR VOLTAGE MAPPING
The unipolar ventricular electrogram, recorded between the tip electrode and an indifferent electrode or Wilson’s central terminal, potentially has a larger “field of view,” and this makes it more susceptible to recording artefacts. Nevertheless, the peak-to-peak amplitude of the initial unipolar forces may allow the effects of deeper layers of fibrosis to be manifest when the bipolar signal is normal. Hutchinson et al found that the mean unipolar left ventricular electrogram voltage recorded from normal ventricles in patients without structural heart disease was 19.6 ± 6.9 mV and 95% of signals were of amplitude >8.27 mV.4 In the right ventricle, with its smaller myocardial mass, the same group was able to define that 95% of normal RV signals have a unipolar voltage >5.5 mV.3
Having defined the ventricle-specific reference values for unipolar voltage in the absence of scar (Table 20-1), these investigators analyzed patients with scar-related VT in the context of left and right ventricular nonischemic cardiomyopathies. Hutchinson et al found that in the absence of endocardial bipolar abnormalities, regions of low endocardial unipolar voltage predicted the presence of overlying epicardial substrate in all patients, with a 61% area overlap.4 Polin et al published similar findings in patients with ARVD in whom endocardial RV free-wall unipolar signal attenuation predicted the presence and extent of overlying epicardial scar in patients with minimal abnormalities of endocardial bipolar signals.3 Hence unipolar endocardial voltage mapping can be used to predict the presence of epicardial substrate, and the potential requirement for obtaining percutaneous pericardial access for VT ablation, in the free walls of both the LV and RV.
TABLE 20-1 Normal Ventricular Electrogram Voltage Criteria
Since unipolar signals may be influenced by the presence of fibrosis deep to the endocardial recording surface, they may also, when they are abnormally attenuated in the absence of overlying epicardial scar, predict the presence of intramural substrate. This is of particular relevance in the group of nonischemic DCM patients who have septal fibrosis. Of DCM patients with monomorphic VT undergoing catheter ablation, around 12% have isolated septal substrate without the usual basolateral fibrosis being present.11 As demonstrated in the case study, these patients sometimes have completely normal endocardial bipolar voltage and a paucity of epicardial substrate, generally only in the preaortic LV summit region. MRI characteristically displays an intramural layer of DGE deep in the ventricular septum. Bipolar biventricular voltage mapping of the septum is not surprisingly normal in such situations due to the normal RV and LV septal endomyocardium overlying the intramural fibrosis. Unipolar mapping should be performed with a normal voltage cut-off of >8.27 mV for both the RV as well as the LV septum due to its greater muscle mass than the RV free wall. Once an area of septal fibrosis is identified, mapping efforts, particularly pace-mapping, can be directed to this region, but ablation of VT associated with septal substrate remains challenging with currently available techniques and energy sources.11
Substrate-based strategies are frequently required for successful ablation of scar-related VT as the majority of patients have unmappable VT morphologies. Voltage mapping underpins these strategies as it is the first and often most crucial step in defining the regions where VT circuit components reside. Depending on the particular form of structural heart disease present, both endocardial and epicardial mapping may be required, and both bipolar and unipolar voltage analysis may be pivotal.
1. Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation. 2000;101(11):1288-1296.
2. Cano O, Hutchinson MD, Lin D, et al. Electroanatomic substrate and ablation outcome for suspected epicardial ventricular tachycardia in left ventricular nonischemic cardiomyopathy. J Am Coll Cardiol. 2009;54(9):799-808.
3. Polin GM, Haqqani HM, Tzou W, et al. Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm. 2011;8(1):76-83.
4. Hutchinson MD, Gerstenfeld EP, Desjardins B, et al. Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Circ Arrhythm Electrophysiol. 2011;4(1):49-55.
5. Cassidy DM, Vassallo JA, Marchlinski FE, Buxton AE, Untereker WJ, Josephson ME. Endocardial mapping in humans in sinus rhythm with normal left ventricles: activation patterns and characteristics of electrograms. Circulation. 1984;70(1):37-42.
6. Bogun F, Good E, Reich S, et al. Isolated potentials during sinus rhythm and pace-mapping within scars as guides for ablation of post-infarction ventricular tachycardia. J Am Coll Cardiol. 2006;47(10):2013-2019.
7. Jais P, Maury P, Khairy P, et al. Elimination of local abnormal ventricular activities: a new end point for substrate modification in patients with scar-related ventricular tachycardia. Circulation. 2012;125(18):2184-2196.
8. Gepstein L, Hayam G, Ben-Haim SA. A novel method for nonfluoroscopic catheter-based electroanatomical mapping of the heart. In vitro and in vivo accuracy results. Circulation. 1997;95(6):1611-1622.
9. Haqqani HM, Marchlinski FE. Electrophysiologic substrate underlying postinfarction ventricular tachycardia: characterization and role in catheter ablation. Heart Rhythm. 2009;6(8):S70-76.
10. Reddy VY, Wrobleski D, Houghtaling C, Josephson ME, Ruskin JN. Combined epicardial and endocardial electroanatomic mapping in a porcine model of healed myocardial infarction. Circulation. 2003(25);107:3236-3242.
11. Haqqani HM, Tschabrunn CM, Tzou WS, et al. Isolated septal substrate for ventricular tachycardia in nonischemic dilated cardiomyopathy: incidence, characterization, and implications. Heart Rhythm. 2011;8(8):1169-1176.