Bipolar Mapping, Revisited
Bipolar Mapping, Revisited
Then I could travel just by folding a map
No more airplanes, or speed trains, or freeways
There'd be no distance that could hold us back."
"The New Year,"
Death Cab for Cutie
The substantial technological advancements over the past decade in the catheter ablation of ventricular arrhythmias (VA) have failed to produce a proportional improvement in procedural outcome. Particularly in patients with nonischemic left ventricular cardiomyopathy (LVCM), the driving force for innovation has rightly focused upon enhanced characterization of arrhythmia substrate. The classical 2-dimensional model for reentrant ventricular tachycardia (VT) in which all of the critical circuit elements are located superficially within the subendocardium is a clear oversimplification when applied to the transmural circuits encountered in many LVCM patients. Historically, the identification of patients with complex substrate was often made in the wake of extensive and ineffective endocardial ablation. Despite widespread adoption of the "brute force" approach to VT ablation in which the entirety of the arrhythmia substrate is targeted, most VT ablationists yearn for a physiologic renaissance. Too often clinical practice is shaped by the latest new technology, instead of refining and maximizing data from readily available sources.
In response to this important diagnostic and therapeutic challenge, many electrocardiographic and imaging techniques have been developed that provide clues to the presence of arrhythmia substrate remote from the endocardium. The need for epicardial ablation, however, is often difficult to determine a priori due to patient- or institution-specific challenges in preoperative data collection (e.g., lack of source electrocardiogram [ECG] data, inability to perform magnetic resonance [MR] imaging in device patients). It is also clear that the mere presence of intramural or epicardial substrate does not necessarily portend failure of endocardial ablation. Thus, the guiding principle of an organized approach to ablation of the patient with VA lies not only in having the tools to optimize the likelihood of success but also in the proper interpretation of the data to help guide the catheter to the "right spot."
In this issue of the Journal, Tzou et al. describe a novel methodology for determining the need for epicardial ablation in a (largely) LVCM population using bipolar electroanatomical mapping characteristics. They found the following criteria predictive of requiring epicardial ablation: (1) a diffusely early endocardial activation pattern, (2) the presence of an endocardial far-field electrogram (EGM) preceding the sharp near-field EGM in the region of early activation, and (3) failure to capture the far-field EGM or replicate the spontaneous VA morphology with overdrive pacing. Their findings contribute substantially to the evolving literature regarding epicardial ablation, and the authors are to be congratulated. However, it is important to take these findings in context and consider their generalizability to the VT ablation population at large.
Several centers have described the value of delayed enhancement (DE) MR in patients with LVCM to define scar transmurality. Furthermore, there has been extensive literature describing the use of the 12-lead ECG in defining site of VT "origin." Although these technologies are indispensable, they have important potential limitations. First, DE MR may lack the spatial resolution to define more limited regions of scar when compared with high-density voltage mapping; it also offers no insight into the site of origin for idiopathic VAs. And although the 12-lead ECG provides clues the exit site or site of origin (for reentrant and focal VT, respectively) of the targeted VA, it does not guarantee that site is the either readily available or safe to ablate. These limitations in part arise from the 3-dimensional natures of arrhythmia circuits within the thick-walled ventricle. This concept is underscored by the failure to replicate with pacemapping the 12-lead ECG morphology of VAs having an intramural origin.
A reduction in endocardial unipolar EGM amplitude, even in the absence of endocardial bipolar scar, has been shown to predict the presence of epicardial scar in both right ventricular (RV) and LVCM patients. This is supported by comparisons with MRI, where endocardial bipolar voltage attenuation has been more strongly correlated with endocardial scar while unipolar voltages afford a more transmural breadth of view. These data suggest that the "field-of-view" of bipolar EGMs may be limited when using a traditional cutoff of 1.5 mV; whether increasing the threshold of bipolar voltage amplitude will provide a similar field-of-view with adequate specificity remains unclear. Although these criteria may suggest the presence of complex substrate patterns, they do not provide physiologic data regarding the 3-dimensional nature of putative VT circuits. In contrast, the current report by Tzou et al. focuses upon EGM characteristics during the spontaneous arrhythmia, thereby refining the discrimination of arrhythmogenic substrate. This distinction is of critical importance when designing the most expedient and safe procedural approach to eliminate the targeted VA.
Interpreting the relevance of EGM morphology during VAs remains an important challenge; often such EGMs display both "sharp" and "dull" deflections with variable frequency content. It is axiomatic that these complex EGMs represent simultaneous recording of both "near-field" and "far-field" myocardium. Unfortunately, the terms "near-field" and "far-field" are deliberately vague and are sparsely described in the literature. It, therefore, seems worthwhile to qualify the terms with respect to the specific context in which they are recorded. Tung et al. previously described criteria to distinguish far-field potentials in patients with postinfarction reentrant VT based upon the observed response to ventricular overdrive pacing. EGMs were considered far-field if they satisfied the following criteria: (1) that there was a multicomponent signal consistently linked to the QRS; (2) that pacing entrained the ventricular arrhythmia (irrespective of the presence of fusion); and (3) that the pacing stimulus did not obscure the EGM. Far-field EGMs were reported in 82% of patients from the same study; however, this has not been similarly characterized in patients with LVCM, in whom the presence of epicardial substrate is more common. Such a schema seems appropriately complex to parse the "Martian" landscape of scar-related VT. In the simpler case of focal VAs originating from the pulmonary artery or aortic root, near- and far-field EGMs are often recorded simultaneously, and their activation pattern reverses in the VA compared with sinus rhythm.
When differentiating near- and far-field EGMs, it is important to consider not only the distance between the working and reference electrodes (i.e., electrode spacing) but also the size of the virtual electrode created during pacing. Both remote capture of adjacent myocardium and anodal stimulation with bipolar pacing of sufficient current density may confound appropriate EGM interpretation. It seems clear that "far-field" is a relative concept. Given the qualitative nature of the far-field EGM analysis advocated in the current study, the interobserver reproducibility of the reported association is unclear.
Furthermore, the ability to record a far-field EGM does not guarantee that it can be eliminated with ablation; this may be particularly relevant in intramural circuits. This issue arises because it is unclear how spatially "remote" such an EGM lies based on its morphology alone. In theory, the pacing output needed to capture the EGM may reflect this distance, though myocardial characteristics, catheter contact, and other factors may strongly impact this observation. The current study would suggest that some critical far-field EGMs recorded from the endocardium are of adequate size or distance from the endocardial breakthrough site to be unaltered by irrigated endocardial ablation.
Although the current study is limited by both its retrospective nature and its lack of pathological correlation for the electrical phenomena described, the findings represent an incremental step toward achieving improved procedural outcomes in patients requiring epicardial ablation to eliminate their VAs. Almost of equal importance to determining the need for epicardial ablation, the authors' findings may help to limit unnecessary and ineffective endocardial ablation in many patients.
The study cohort clearly represents an ideal subset of (largely LVCM) patients with epicardial VA origin, and in whom a diffuse area of endocardial breakthrough was present. These patients had VAs that were either spontaneously occurring or inducible, and were amenable hemodynamically to activation mapping. The study observations may not apply to patients with significant endocardial/intramural scar or multiple endocardial breakthroughs. Furthermore, EGM assessment may depend upon several factors including: (1) the recording electrode size, interelecrode spacing, and catheter tip orientation; (2) the EGM filtering and recording platform utilized; and (3) the presence of external noise or recording artifact. Also, the proposed interpretation of EGM characteristics is inherently subjective, and prospective validation of these criteria is needed to ensure their consistent application across centers.
Another limitation in the current study is that all patients had ECG criteria suggesting an epicardial exit for their VA. Thus, whether the proposed criteria have incremental value in guiding the timing of an epicardial approach had such ECG criteria been unmet remains unclear. Since the VT circuits in LVCM are often transmural, endocardial ablation may effectively eliminate some reentrant VTs exiting on the epicardium. As previously stated, the presence of an epicardial VA exit site/origin guarantees neither a safe nor an effective site to ablate.
Editorial Comment
" I wish the world was flat like the old daysThen I could travel just by folding a map
No more airplanes, or speed trains, or freeways
There'd be no distance that could hold us back."
"The New Year,"
Death Cab for Cutie
The substantial technological advancements over the past decade in the catheter ablation of ventricular arrhythmias (VA) have failed to produce a proportional improvement in procedural outcome. Particularly in patients with nonischemic left ventricular cardiomyopathy (LVCM), the driving force for innovation has rightly focused upon enhanced characterization of arrhythmia substrate. The classical 2-dimensional model for reentrant ventricular tachycardia (VT) in which all of the critical circuit elements are located superficially within the subendocardium is a clear oversimplification when applied to the transmural circuits encountered in many LVCM patients. Historically, the identification of patients with complex substrate was often made in the wake of extensive and ineffective endocardial ablation. Despite widespread adoption of the "brute force" approach to VT ablation in which the entirety of the arrhythmia substrate is targeted, most VT ablationists yearn for a physiologic renaissance. Too often clinical practice is shaped by the latest new technology, instead of refining and maximizing data from readily available sources.
In response to this important diagnostic and therapeutic challenge, many electrocardiographic and imaging techniques have been developed that provide clues to the presence of arrhythmia substrate remote from the endocardium. The need for epicardial ablation, however, is often difficult to determine a priori due to patient- or institution-specific challenges in preoperative data collection (e.g., lack of source electrocardiogram [ECG] data, inability to perform magnetic resonance [MR] imaging in device patients). It is also clear that the mere presence of intramural or epicardial substrate does not necessarily portend failure of endocardial ablation. Thus, the guiding principle of an organized approach to ablation of the patient with VA lies not only in having the tools to optimize the likelihood of success but also in the proper interpretation of the data to help guide the catheter to the "right spot."
In this issue of the Journal, Tzou et al. describe a novel methodology for determining the need for epicardial ablation in a (largely) LVCM population using bipolar electroanatomical mapping characteristics. They found the following criteria predictive of requiring epicardial ablation: (1) a diffusely early endocardial activation pattern, (2) the presence of an endocardial far-field electrogram (EGM) preceding the sharp near-field EGM in the region of early activation, and (3) failure to capture the far-field EGM or replicate the spontaneous VA morphology with overdrive pacing. Their findings contribute substantially to the evolving literature regarding epicardial ablation, and the authors are to be congratulated. However, it is important to take these findings in context and consider their generalizability to the VT ablation population at large.
All Scars are not Created Equal
Several centers have described the value of delayed enhancement (DE) MR in patients with LVCM to define scar transmurality. Furthermore, there has been extensive literature describing the use of the 12-lead ECG in defining site of VT "origin." Although these technologies are indispensable, they have important potential limitations. First, DE MR may lack the spatial resolution to define more limited regions of scar when compared with high-density voltage mapping; it also offers no insight into the site of origin for idiopathic VAs. And although the 12-lead ECG provides clues the exit site or site of origin (for reentrant and focal VT, respectively) of the targeted VA, it does not guarantee that site is the either readily available or safe to ablate. These limitations in part arise from the 3-dimensional natures of arrhythmia circuits within the thick-walled ventricle. This concept is underscored by the failure to replicate with pacemapping the 12-lead ECG morphology of VAs having an intramural origin.
A reduction in endocardial unipolar EGM amplitude, even in the absence of endocardial bipolar scar, has been shown to predict the presence of epicardial scar in both right ventricular (RV) and LVCM patients. This is supported by comparisons with MRI, where endocardial bipolar voltage attenuation has been more strongly correlated with endocardial scar while unipolar voltages afford a more transmural breadth of view. These data suggest that the "field-of-view" of bipolar EGMs may be limited when using a traditional cutoff of 1.5 mV; whether increasing the threshold of bipolar voltage amplitude will provide a similar field-of-view with adequate specificity remains unclear. Although these criteria may suggest the presence of complex substrate patterns, they do not provide physiologic data regarding the 3-dimensional nature of putative VT circuits. In contrast, the current report by Tzou et al. focuses upon EGM characteristics during the spontaneous arrhythmia, thereby refining the discrimination of arrhythmogenic substrate. This distinction is of critical importance when designing the most expedient and safe procedural approach to eliminate the targeted VA.
The Relativity of "Far"
Interpreting the relevance of EGM morphology during VAs remains an important challenge; often such EGMs display both "sharp" and "dull" deflections with variable frequency content. It is axiomatic that these complex EGMs represent simultaneous recording of both "near-field" and "far-field" myocardium. Unfortunately, the terms "near-field" and "far-field" are deliberately vague and are sparsely described in the literature. It, therefore, seems worthwhile to qualify the terms with respect to the specific context in which they are recorded. Tung et al. previously described criteria to distinguish far-field potentials in patients with postinfarction reentrant VT based upon the observed response to ventricular overdrive pacing. EGMs were considered far-field if they satisfied the following criteria: (1) that there was a multicomponent signal consistently linked to the QRS; (2) that pacing entrained the ventricular arrhythmia (irrespective of the presence of fusion); and (3) that the pacing stimulus did not obscure the EGM. Far-field EGMs were reported in 82% of patients from the same study; however, this has not been similarly characterized in patients with LVCM, in whom the presence of epicardial substrate is more common. Such a schema seems appropriately complex to parse the "Martian" landscape of scar-related VT. In the simpler case of focal VAs originating from the pulmonary artery or aortic root, near- and far-field EGMs are often recorded simultaneously, and their activation pattern reverses in the VA compared with sinus rhythm.
When differentiating near- and far-field EGMs, it is important to consider not only the distance between the working and reference electrodes (i.e., electrode spacing) but also the size of the virtual electrode created during pacing. Both remote capture of adjacent myocardium and anodal stimulation with bipolar pacing of sufficient current density may confound appropriate EGM interpretation. It seems clear that "far-field" is a relative concept. Given the qualitative nature of the far-field EGM analysis advocated in the current study, the interobserver reproducibility of the reported association is unclear.
Furthermore, the ability to record a far-field EGM does not guarantee that it can be eliminated with ablation; this may be particularly relevant in intramural circuits. This issue arises because it is unclear how spatially "remote" such an EGM lies based on its morphology alone. In theory, the pacing output needed to capture the EGM may reflect this distance, though myocardial characteristics, catheter contact, and other factors may strongly impact this observation. The current study would suggest that some critical far-field EGMs recorded from the endocardium are of adequate size or distance from the endocardial breakthrough site to be unaltered by irrigated endocardial ablation.
Tempered Enthusiasm
Although the current study is limited by both its retrospective nature and its lack of pathological correlation for the electrical phenomena described, the findings represent an incremental step toward achieving improved procedural outcomes in patients requiring epicardial ablation to eliminate their VAs. Almost of equal importance to determining the need for epicardial ablation, the authors' findings may help to limit unnecessary and ineffective endocardial ablation in many patients.
The study cohort clearly represents an ideal subset of (largely LVCM) patients with epicardial VA origin, and in whom a diffuse area of endocardial breakthrough was present. These patients had VAs that were either spontaneously occurring or inducible, and were amenable hemodynamically to activation mapping. The study observations may not apply to patients with significant endocardial/intramural scar or multiple endocardial breakthroughs. Furthermore, EGM assessment may depend upon several factors including: (1) the recording electrode size, interelecrode spacing, and catheter tip orientation; (2) the EGM filtering and recording platform utilized; and (3) the presence of external noise or recording artifact. Also, the proposed interpretation of EGM characteristics is inherently subjective, and prospective validation of these criteria is needed to ensure their consistent application across centers.
Another limitation in the current study is that all patients had ECG criteria suggesting an epicardial exit for their VA. Thus, whether the proposed criteria have incremental value in guiding the timing of an epicardial approach had such ECG criteria been unmet remains unclear. Since the VT circuits in LVCM are often transmural, endocardial ablation may effectively eliminate some reentrant VTs exiting on the epicardium. As previously stated, the presence of an epicardial VA exit site/origin guarantees neither a safe nor an effective site to ablate.