Implantable cardioverter defibrillator therapy can prevent. Original Article

Transcription

1 Original Article Electroanatomical Voltage and Morphology Characteristics in Postinfarction Patients Undergoing Ventricular Tachycardia Ablation Pragmatic Approach Favoring Late Potentials Abolition Dimitris Tsiachris, MD; John Silberbauer, MD; Giuseppe Maccabelli, MD; Teresa Oloriz, MD; Francesca Baratto, MD; Hiroya Mizuno, MD; Caterina Bisceglia, MD; Pasquale Vergara, MD; Alessandra Marzi, MD; Nicoleta Sora, MD; Fabrizio Guarracini, MD; Andrea Radinovic, MD; Manuela Cireddu, MD; Simone Sala, MD; Simone Gulletta, MD; Gabriele Paglino, MD; Patrizio Mazzone, MD; Nicola Trevisi, MD; Paolo Della Bella, MD Background Catheter ablation is an important therapeutic option in postmyocardial infarction patients with ventricular tachycardia (VT). We analyzed the endo epicardial electroanatomical mapping (EAM) voltage and morphology characteristics, their association with clinical data and their prognostic value in a large cohort of postmyocardial infarction patients. Methods and Results We performed total and segmental analysis of voltage (bipolar dense scar [DS] and low voltage areas, unipolar low voltage and penumbra areas) and morphology characteristics (presence of abnormal late potentials [LPs] and early potentials [EPs]) in 100 postmyocardial infarction patients undergoing electroanatomical mapping based VT ablation (26 endo epicardial procedures) from All patients had unipolar low voltage areas, whereas 18% had no identifiable endocardial bipolar DS areas. Endocardial bipolar DS area >22.5 cm 2 best predicted scar transmurality. Endo epicardial LPs were recorded in 2/3 patients, more frequently in nonseptal myocardial segments and were abolished in 51%. Endocardial bipolar DS area >7 cm 2 and endocardial bipolar scar density >0.35 predicted epicardial LPs. Isolated LPs are located mainly epicardially and EPs endocardially. As a primary strategy, LPs and VT-mapping ablation occurred in 48%, only VT-mapping ablation in 27%, only LPs ablation in 17%, and EPs ablation in 6%. Endocardial LP abolition was associated with reduced VT recurrence and increased unipolar penumbra area predicted cardiac death. Conclusions Endocardial scar extension and density predict scar transmurality and endo epicardial presence of LPs, although DS is not always identified in postmyocardial infarction patients. LPs, most frequently located in nonseptal myocardial segments, were abolished in 51% resulting in improved outcome. (Circ Arrhythm Electrophysiol. 2015;8: DOI: /CIRCEP ) Key Words: catheter ablation endocardium myocardium myocardial infarction tachycardia, ventricular Implantable cardioverter defibrillator therapy can prevent sudden cardiac death because of sustained ventricular tachycardia (VT) late after myocardial infarction (MI). 1 However, patients experiencing implantable cardioverter defibrillator shocks have a decreased quality of life and increased mortality compared with patients without shocks. 2,3 Catheter ablation is an important therapeutic option in patients with VT late after MI. 4,5 The majority of inducible VTs in these patients are unmappable. 6 A variety of alternative substrate-based methods have evolved to overcome the shortcomings of activation and entrainment mapping Substrate ablation during sinus rhythm consists of late potential (LP) abolition, elimination of local abnormal ventricular activities (LAVAs), conduction channel ablation, linear isthmus ablation, electric isolation of low voltage area, and scar homogenization, all identified by electroanatomical mapping (EAM) The purpose of this study was to analyze the endo epicardial EAM voltage and morphology characteristics to describe the appropriateness of each substrate ablation strategy, the interrelationship of EAM characteristics, their association with clinical data, and their prognostic value in a large cohort of post-mi patients undergoing EAM-based catheter ablation for VT. Received October 27, 2015; accepted May 13, From the Arrhythmia Unit and Electrophysiology Laboratories, Ospedale San Raffaele, Milan, Italy. The Data Supplement is available at Correspondence to Paolo Della Bella, MD, Arrhythmia Unit and Electrophysiology Laboratories, Ospedale San Raffaele, Via Olgettina 60, Milan, Italy. dellabella.paolo@hsr.it 2015 American Heart Association, Inc. Circ Arrhythm Electrophysiol is available at DOI: /CIRCEP

2 864 Circ Arrhythm Electrophysiol August 2015 WHAT IS KNOWN For ventricular tachycardia because of prior myocardial infarction, substrate ablation targets detected during sinus rhythm may include late and early potentials, conduction channels. The relation of these markers to areas with low bipolar and unipolar voltage representing scars is not completely characterized. WHAT THE STUDY ADDS Late potentials are present in two-thirds of patients associated with large and more confluent appearing low voltage regions and are most frequently located in nonseptal myocardial segments. Endocardial scar characteristics predict also the existence of potential arrhythmogenic substrate in the epicardium. Abolition of late potentials is feasible in half of patients is associated with fewer recurrences rate and we suggest is the most appropriate substrate ablation strategy in post-myocardial infarction patients. Methods Study Population One hundred sixty consecutive patients with post-mi drug-refractory VT were referred for catheter ablation at the VT Unit at the San Raffaele Hospital, Milan, between January 2010 and December Of these, 60 patients were excluded as they had a non-carto 3 (Biosense-Webster, Diamond Bar, CA) based procedure. The study therefore included 100 patients with analyzable Carto 3 substrate maps that provide local activation time and both bipolar ( Hz) and unipolar (1 240 Hz) voltage data. The study period ended in December 2013 allowing at least 6 months of follow-up in every patient. The study was approved by the institutional review committee, and all patients gave written informed consent. VT Ablation Set-Up Details about the status of coronary artery disease and presence of comorbidities were recorded in all patients. The diagnosis and location of MI were established by the history, pathological Q waves on ECG, regional wall-motion abnormalities, or perfusion defects on imaging that correlated with coronary angiography (>70% stenosis of an epicardial coronary artery). Arrhythmia presentation was classified as either electrical storm, incessant, or paroxysmal VT as previously described. 13 Amiodarone was stopped 48 hours before the procedure. Procedures were performed under general anesthesia. Subxiphoid epicardial access was undertaken after a prior failed endocardial procedure if an epicardial circuit was suspected (n=16) or as protocol-guided first-line approach (n=10). In the latter cases, no clinical data or a prior failed procedure indicated the need for epicardial procedure. Double left ventricular (LV) access (retrograde aortic and transseptal) was standard. The transseptal sheath was left heparinized in the left atrium when not used. Electroanatomical Mapping Voltage Analysis High-density substrate mapping with a 5-mm fill threshold in low bipolar voltage areas (LVA) and 10 mm elsewhere was performed using the Carto 3 workstation and a 3.5-mm open irrigated catheter with contact force assessment since available (Navistar Thermocool, Biosense-Webster, Diamond Bar, CA). In patients with ventricular pacing, EAM was performed during spontaneous rhythm, if feasible. LV endo and epicardial bipolar dense scar (DS) was defined as an area with a bipolar voltage 0.5 mv and areas with a bipolar voltage >0.5 and <1.5 mv were defined as bipolar border zone (Figure 1). Endo and epicardial unipolar scar were areas with a unipolar voltage <8 mv. 14 Area measurements were achieved using the area measurement tool. Endo and epicardial bipolar scar density was defined as the ratio of the bipolar DS area to total LVA, an index assumed to reflect the density of the myocardial fibrosis within the infarct region. Matching DS identified on the endocardium and epicardium was presumed to be transmural scar. We defined as unipolar penumbra area the unipolar scar beyond the bipolar LVA. 15 Further offline analyses were undertaken after segmenting the LV endocardial shell into 17 segments and the LV epicardial shell into 12 segments using the design line tool. 16 Each segment was considered to be analyzable only if fill threshold criteria were met and the predominant bipolar and unipolar voltage type was noted. Electrogram Analysis A color-coded map of sinus rhythm activation delay was drawn on the same anatomic shell by manual tagging the latest activity of all electrograms (LPs map). The definition of LPs included either continuous fragmented activity bridging from the main component within the QRS to the latest signal recorded outside the QRS, without a definite voltage cutoff (fractionated LPs), or isolated potentials recorded after the QRS offset (isolated LPs). Baseline LPs maps were used to define the localization and size of LP areas and were compared with remaps created post ablation (Figure 1). Early potentials (EPs) were defined as fractionated (EGM containing >4 sharp deflections) or isolated ( 2 sharp EGMs separated by an isolelectric segment) within the QRS; pacing was systematically attempted at these sites looking for morphology match with any induced VT and for latency between the stimulus and the QRS (>40 ms). Pacing was undertaken at just above the capture threshold aiming for near-field capture only. The predominant electrogram (EGM) type was also noted for each segment. LP abolition was a primary target in all patients and ablation of selected EPs secondary or adjunctive. VT Ablation Strategy After high-density substrate mapping, programmed ventricular stimulation with 4 extrastimuli from the right ventricular apex and multiple LV sites was performed and repeated at the end of the procedure. By study design, we aimed to achieve the combined procedural end point of VT noninducibility and LP abolition in all cases. Our VT ablation strategy is outlined: 1. In patients with tolerated VT, VT was ablated using activation and entrainment mapping. Ablation continued in sinus rhythm aiming at complete abolition of LPs when present or EPs when indicated. 2. In patients with noninducible or hemodynamically nontolerated VT, ablation was performed during sinus rhythm, targeting LP areas when present and EPs when indicated. For each segment, the reason for ablation was also recorded including 1 or more of the following: VT mapping, LPs, and EPs. Follow-Up In patients with a successful ablation, prior amiodarone was not reinstituted and β-blocker therapy maintained. VTs were mostly recorded through home-monitoring while outpatient follow-up visits were scheduled at 3 months and then 6-monthly intervals or whenever symptoms recurred. Primary study end-points were VT recurrence and occurrence of cardiac death.

3 Tsiachris et al EAM Features in Post-MI VT Ablation 865 Figure 1. See the next page for the full figure legend.

4 866 Circ Arrhythm Electrophysiol August 2015 Figure 1 (Continued). Anteroposterior view of a large anteroapical epicardial (A) and endocardial (B) bipolar scar (A and B, red color is dense scar; purple is normal; and the yellow-green area is low voltage). A large endocardial dense scar predicts the presence of epicardial dense scar and areas of late potentials epicardially (C and D) and endocardially (E and F). In C F, red color is within QRS electrogram and purple is late potential. Isolated late potentials are more often found in the epicardium (D) and fractionated late potentials in the endocardium (F). The remap (G J) verifies successful late potentials abolition epicardially (H) and endocardially (J). Statistical Analysis We analyzed EAM voltage and EGM characteristics separately in the endocardium and in the epicardium, focusing on the predictors of epicardial presence of DS and LPs by constructing receiver operating characteristic curves. We also related clinical and EAM characteristics in patients with LPs compared with those without LPs (even according to infarct site) and we conducted segmental voltage and EGM analysis aiming to detect differences in the endocardium compared with the epicardium and furthermore dividing the LV shell into anterior, septal, and inferolateral parts. Continuous variables were presented as either means (±SD) or medians (with Q1 Q3) and categorical variables as numbers and percentages. Comparisons between groups were undertaken using the t test or nonparametric tests for continuous variables and Fisher s exact or χ 2 tests for proportions as indicated. Analysis of ablation characteristics along with the above findings may help assessing the appropriateness of different substrate ablation strategies. Cox proportional hazards analyses were used to assess the prognostic role only of endocardial EAM characteristics building clinical rather than statistical models. Differences were considered statistically significant at the 2-sided P<0.05 level. All statistical analyses were performed using the SPSS version 15.0 statistical software (SPSS Inc, Texas, IL). Results No difference was observed in any of the background characteristics between the enrolled (n=100) and excluded (n=60) patients (Table I in the Data Supplement). Of the 100 included patients, 36 patients had an anterior MI and 64 had an inferolateral MI with 17 having had at least 1 prior failed catheter ablation procedure. Among inferolateral infarcts, right coronary artery was the culprit vessel in 43 patients, left circumflex in 17 patients, and both right coronary artery and left circumflex in 4 patients. Scar Analysis The mean endocardial surface area was 236.1±58 cm 2. Of that, 10.2% was bipolar DS, 21.8% was bipolar LVA, and 46.7% was unipolar scar. Endocardial penumbra was present in all but 1 patient, and mean area was 51.6 cm 2 (24.9%). Although unipolar scar was present in all cases, 18% of the post-mi patients had no evidence of endocardial bipolar DS. In 14 of these, bipolar LVA was recorded and endocardial LPs were present in 5. All these 18 patients had a history of MI (9 anterior and 9 inferolateral), regional wall motion abnormalities (mean LVEF=37%), and established coronary artery disease (8 had 3-vessel and 7 had 2-vessel disease), whereas 13 of them were revascularized (5 with CABG and 8 with PCI). Among the 26 patients with available epicardial maps, 8 had no bipolar DS and 4 neither LVA nor DS. Epicardial penumbra was absent in 5/26 patients. Based on receiver operating characteristic curve analyses, endocardial bipolar DS area was the only predictor of the epicardial presence of bipolar DS (area under the curve, 0.75; confidence intervals, ; P=0.040) with an optimal value of 22.5 cm 2 (sensitivity, 61.1% and specificity, 87.5%). Endocardial penumbra area had no predictive value for the epicardial presence of either DS or LVA. Segmental Scar Analysis Within the study population, 1597 of 1700 endocardial segments fulfilled fill threshold criteria and were included in the EGM analysis. DS was the predominant bipolar voltage type in 14.3% of the segments and border zone in 20%. Unipolar scar was present in 60.1%. In all bipolar DS segments, there was also unipolar scar. In contrast, in 23.9% of unipolar scar segments, there was underlying predominant bipolar DS, in 32.3% bipolar border zone, and in 43.8% endocardial penumbra (normal bipolar amplitude). Presence of LPs Endocardial LPs were recorded in 66% (n=66) of the patients and epicardially in 17/26 (65.4%). Prevalence of epicardial LPs was not different between research protocol patients and those with a prior failed procedure as the indication for epicardial mapping (80% versus 56%; P=0.21). Abolition of endocardial LPs was achieved in 51/66 (77.3%) and of epicardial LPs in 10/17 (58.8%) patients. We identified 3 patients in whom LPs were abolished endocardially and persisted epicardially. Of them, 1 remained inducible at the end of the procedure. Patients with endocardial LPs had greater endocardial bipolar DS (P<0.001), bipolar LVA (P=0.001), and ablation area (P=0.029), as well as increased endocardial bipolar scar density (P<0.001), compared with those without LPs (Table 1). Similarly, patients with epicardial LPs compared with those without exhibited increased endocardial bipolar DS area (P<0.001) and bipolar scar density (P=0.006), as well as epicardial bipolar DS (P=0.008), bipolar LVA (P=0.011), unipolar scar areas (P=0.001), and bipolar scar density (P=0.034), despite the decreased power to detect a difference based on the small number of patients in each group (Table 2; Table II in the Data Supplement). Accordingly, there was no difference in either endocardial or epicardial penumbra areas according to the presence of endocardial or epicardials LPs. Based on receiver operating characteristic curve analyses, endocardial bipolar DS area (area under the curve, 0.80; confidence intervals ; P=0.011) and endocardial bipolar scar density (area under the curve, 0.82; confidence intervals, ; P=0.008) predicted the epicardial presence of LPs with optimal values of 7 cm 2 (sensitivity, 88.2% and specificity, 66.7%) and 0.35 (sensitivity, 52.9% and specificity, 100%), respectively. Notable, a value of endocardial bipolar DS area >38 cm 2 had 100% specificity (Figure 2). Endocardial penumbra did not predict the epicardial presence of LPs. Segmental EGM Analysis LPs were predominantly present in 10% of the 1597 endocardial segments (in 5.2% fractionated LPs and in 4.8% isolated LPs), EPs in 22.1% (in 19.3% fractionated EPs and in 2.8% isolated EPs), and normal EGMs in 66.8% of the segments (no EGM type was identified in 1.1% of the segments because of diffuse DS). Unipolar scar was present in all segments with

5 Tsiachris et al EAM Features in Post-MI VT Ablation 867 Table 1. Procedural and Mapping Characteristics According to Endocardial LP Presence or Not Endocardial Presence of LPs (n=66) Endocardial Absence of LPs (n=34) P Value Background Anterior myocardial infarction, % Left ventricular ejection fraction, % 31.8± ± Procedural data Epicardial access, % Procedure time, min 243.1± ± * VTs in procedure ( ) 2 (1 3) 1 (1 3) 0.55 Nontolerated VT VTs terminated ( ) 1 (0 1.2) 1 (0 1) 0.72 Electroanatomical data ENDO surface area, cm ± ± ENDO points 402.5± ± ENDO LV volume, cm ± ± ENDO BIP dense scar area, cm 2 23 ( ) 4.5 ( ) <0.001* ENDO BIP low voltage area, cm ( ) 25 (8 59.7) 0.001* ENDO absence of BIP dense scar, % <0.001* ENDO BIP scar density 0.32 ( ) 0.14 ( ) <0.001* ENDO unipolar scar area, cm ( ) 100 ( ) ENDO penumbra area, cm ( ) 47.5 ( ) 0.73 Ablation area, cm ± ± * Ablation time, min 28.9± ± VT end, % BIP indicates bipolar; ENDO, endocardial; LPs, late potentials; LV, left ventricular; and VT, ventricular tachycardia. *P<0.05 (statistical significance). LPs. Underlying bipolar DS was present in 60% of segments with LPs, border zone in 33.8%, and endocardial penumbra in 6.2% (normal bipolar amplitude). In 98.3% of segments with EPs, there was underlying unipolar scar. Bipolar DS was present in 32.3% of segments with EPs and border zone in 67.7%. LPs were also predominantly present in 12.3% of the epicardial segments (in 2.2% fractionated LPs and in 10.1% isolated LPs), EPs in 13.5% (in 9.4% fractionated EPs and in 4.1% isolated EPs), normal EGMs in 71.9%, and diffuse DS in 2.2%. Importantly, among the 12.3% of epicardial segments with LPs, isolated LPs were found in 10.1% (82.1%), whereas this percent was 48% with respect to the epicardium. Focusing on the distribution of LPs in the 17-segment shell, we exhibited a significant shift toward the inferolateral segments taking into account the whole study population (Figure 3). Analysis According to Infarct Site Patients with anterior infarcts (n=36) compared with those with inferolateral infarcts (n=64) were younger ( 3.8 years, P=0.021), had lower LVEF ( 3.9%, P=0.049), a greater prevalence of LV aneurysm (+28%, P<0.001) and exhibited increased total endocardial surface area (P=0.012), endocardial LV volume (P=0.001), and unipolar scar (P<0.001; Table 3). Patients with anterior compared with those with inferolateral MIs had no difference in the proportion of segments with bipolar DS (14.5% versus 14.2%; P=0.8) and segments with LPs (8.7% versus 10.9%; P=0.07). LV segments analysis was also performed after dividing LV shell into anterior (4/17), septal (5/17), and inferolateral (8/17) segments. DS was present in 26.8% of the anterior segments, in 16% of the septal, and in 7.4% of the inferolateral segments in patients with anterior MIs. Accordingly, LPs were present in 20.7% of the anterior segments, in 5% of the septal (5%), and in 4.9% of the inferolateral. Regarding patients with inferolateral MIs, DS was found in 22.5% of the inferolateral, in 9.3% of the septal, and in 2.9% of the anterior segments. Likewise, LPs were present in 19.3% of the inferolateral, in 4.9% of the septal (4.9%), and in 0.8% of the anterior (Figure 4). Ablation Characteristics At baseline monomorphic VT was induced during PVS in 61 patients, during creation of the substrate map in 14 patients, whereas 11 had baseline incessant VT. The median number of VTs in each procedure was 2 and at least 1 nontolerated VT was present in 38 patients. Ablation was performed during activation mapping in 75% of the patients. This includes also limited mapping of nontolerated VTs when the ablation catheter was placed in the area of latest activity during sinus rhythm, showing mid-diastolic potential during VT. LPs were targeted in all patients with areas of LPs (2/3 of the study population). EPs were targeted in 23% of the patients whenever there was latency (stimulus to QRS >40ms) or morphology matched with induced VTs during pacemapping maneuvers.

6 868 Circ Arrhythm Electrophysiol August 2015 Table 2. Procedural and Mapping Characteristics According to Epicardial LP Presence or Not Focusing on the primary ablation strategy, 48% of the patients underwent both ablation of LPs and VT mapping, and 27% underwent only VT mapping in the absence of LPs. In 17% of the study population, no VT mapping was available and ablation of LPs constituted the principal ablation strategy. Ablation of EPs was the basic strategy in 6% of the post-mi patients with absence of VT mapping and LPs. VT Recurrence No patients were lost to follow-up with a median follow-up time of 628 ( ) days. Thirty-two (32%) had a VT recurrence during the follow-up period with a median recurrence time of 73 (14 217) days. Focusing on the prognostic value only of EAM characteristics, endocardial LP presence (hazard ratio [HR], 0.398; P=0.009) and LP abolition (HR, 0.260; P=0.001) were associated with reduced VT recurrence. In the multivariate analysis with the above 2 EAM characteristics as the sole selected variables, LP abolition was the only predictor of reduced VT recurrence (HR, 0.274; P=0.010). Endo Epicardial Procedures Epicardial Presence of LPs (n=17) Epicardial Absence of LPs (n=9) P Value Background Age 70.7± ± Anterior myocardial infarction, % LV ejection fraction, % 29.2± ± Procedural data Procedure time, min 306.7± ± VTs in procedure ( ) 2 (0.5 3) 3 (1 3) 0.87 Nontolerated VT Electroanatomical data ENDO surface area, cm ± ± ENDO points 356.6± ± ENDO LV volume, cm ± ± ENDO BIP dense scar area, cm 2 33 (11 52) 5 (0 21.5) 0.009* ENDO BIP low voltage area, cm 2 55 ( ) 19 (3 67) ENDO absence of BIP dense scar ENDO BIP scar density 0.36 ( ) 0.11 ( ) 0.006* ENDO unipolar scar area, cm ( ) 85 ( ) 0.22 ENDO penumbra area, cm 2 45 ( ) 52 ( ) 0.71 ENDO LP area, cm ( ) 8 (4 20.2) ENDO LPs, % * EPI BIP dense scar area, cm 2 18 (5-53) 1 (0 12.5) 0.008* EPI BIP low voltage area, cm 2 45 ( ) 14 (0 27.5) 0.011* EPI BIP scar density 0.19 ( ) 0.07 ( ) 0.034* EPI unipolar scar area, cm 2 99 ( ) 20 (6 67.5) 0.001* EPI penumbra area, cm 2 32 ( ) 7 (0 40) ENDO ablation area, cm 2 25 (14 35) 22.6± ENDO ablation time, min 26.4± ± EPI ablation area, cm 2 10 (5.5 21) 0 (0 4.5) <0.001* EPI ablation time, min 11 (6 22.5) 0 (0 5) 0.001* BIP indicates bipolar; ENDO, endocardial; EPI, epicardial; LPs, late potentials; LV, left ventricular; and VT, ventricular tachycardia. *P<0.05 (statistical significance). Cardiac Death Five patients died from noncardiac and 7 (7%) from cardiac causes during the study period (6 because of cardiac decompensation and 1 sudden death). Among EAM characteristics, endocardial LP presence (HR, 0.204; P=0.047), endocardial penumbra area (HR, 1.018; P=0.001), and unipolar scar (HR=1.013, P=0.034) were related with cardiac death. In the multivariate analysis with the above 3 EAM characteristics as the sole selected variables, endocardial LP presence (HR, 0.177; P=0.041) acted as a prophylactic predictor, whereas increased endocardial penumbra area (HR, 1.028; P=0.044) as an adverse predictor of cardiac death. Discussion In this study, we analyzed the EAM voltage and morphology characteristics in post-mi patients undergoing catheter ablation for VT focusing on the identification and ablation of LPs as a principal procedural end point. We concluded that LPs are present in two thirds of post-mi patients with

7 Tsiachris et al EAM Features in Post-MI VT Ablation 869 Figure 2. Based on receiver operating characteristic curve analyses, endocardial bipolar dense scar (DS) area predicted the epicardial presence of both DS and late potentials (LPs) with optimal values of 22.5 cm 2 (sensitivity, 61.1% and specificity, 87.5%) and 7 cm 2 (sensitivity, 88.2% and specificity, 66.7%), respectively. larger and more solid scars, and they are most frequently located in nonseptal myocardial segments. The above endocardial EAM features predict also the existence of a potential arrhythmogenic substrate in the epicardium. Abolition of LPs is feasible in half cases and is associated with lower VT recurrence rate. Evolution of Substrate-Based Ablation Strategies Inherent limitations of VT inducibility as an ablation target, such as effects of antiarrhythmic/sedation/general anesthetic therapy, stimulation protocol pacing site dependence, and inconsistent clinical/nonclinical VT/VF induction stressed out the need for systematic studies of the substrate. 9,17 Based on the surgical experience, early attempts targeted scar border zone creating linear ablation lines through the DS until reaching the normal myocardium or valve continuity. 6 Further research focused on identification of critical VT isthmuses and conducting channels between unexcitable scar areas. 7,8 More recent substrate ablation strategies focused on EGM characteristics beyond scar identification. In this setting, we have consistently presented that complete LPs abolition based on remapping, in combination with VT noninducibility, constitutes an excellent procedural end-point. 9,17 Likewise, the elimination of LAVAs, defined by us as LPs and EPs, has also been associated with a better outcome. 10 These approaches require the ablation of all substrates, not only the substrate acting as a VT isthmus at the moment of the procedure. Toward this direction, the more aggressive in terms of the ablation and less electrophysiologically oriented endo epicardial scar homogenization technique has provided promising results. 11 However, there have been attempts to minimize the required ablation lesions either by targeting the abnormal EGMs with the shortest delay between the far-field component and the delayed local component (eliminating neighboring and remote areas of slow conduction) or by electric isolation of the scar area (<1.5 mv) by means of linear ablations encircling the scar Scar Characteristics in Post-MI Patients In this study, we used the gold standards of EAM cutoff values derived from studies that validated EAM data according to late-enhanced scar in MRI. 21,22 Surprisingly, we identified a significant, previously not recognized, proportion of patients without evident endocardial bipolar DS, despite fulfilling the Figure 3. Dense scar was the predominant bipolar voltage type in 14.3% and late potentials (LPs) in 10% of the endocardial segments. In the 17-segment shell, dense scar is distributed more often in the inferoseptal and inferolateral segments, whereas a shift in the distribution of LPs is observed in the inferolateral segments.

8 870 Circ Arrhythm Electrophysiol August 2015 Table 3. Presentation Characteristics According to Infarct Site Anterior (n=36) Inferior (n=64) P Value Background Men, % Age 68± ± * LV ejection fraction, % 29.8± ± * Chronic renal disease, % Atrial fibrillation, % LV aneurysm, % <0.001* Revascularization, % 0.69 None CABG PCI CABG and PCI Implantable cardioverter defibrillator, % None Single chamber Dual chamber Cardiac resynchronization therapy-defibrillator Presentation, % Electric storm current Electric storm history Incessant VT Shocks Symptomatic VT Procedural data Epicardial Procedure time, min 233.9± ± VTs in procedure ( ) 2 (1 3) 2 (1 3) 0.65 Nontolerated VT Electroanatomical data ENDO surface area, cm ± ± * ENDO points 365.1± ± ENDO LV volume, cm ± ± * ENDO BIP dense scar area, cm 2 20 (2.2 58) 16.5 ( ) 0.44 ENDO BIP low voltage area, cm 2 60 ( ) 47 ( ) 0.07 ENDO unipolar scar area, cm ( ) 87.5 ( ) <0.001* ENDO LPs, % ENDO LP area, cm 2 15 (7.5 21) 16.5 (8.5 27) 0.44 Epicardial LPs, % ENDO ablation area, cm ± ± ENDO ablation time, min 27.3± ± BIP indicates bipolar; CABG, coronary artery bypass grafting; ENDO, endocardial; EPI, epicardial; LPs, late potentials; LV, left ventricular; PCI, percutaneous coronary intervention; and VT, ventricular tachycardia. *P<0.05 (statistical significance). diagnostic criteria of ischemic cardiomyopathy. Indeed, MRI scar areas have been exhibited to be larger in one third of post-mi patients, and visual analysis has showed a clear mismatch between EAM bipolar and MRI scar areas. 23 Beyond the under-recognition of EAM scar, absence of DS could be attributed to the underlying presence of either hibernated myocardium, early reperfused myocardium, remarkable collateral circulation, or the presence of concomitant cardiomyopathy. 24 LPs were found in only 6 of 18 patients with absent bipolar DS and EPs in additional 3 patients, limiting significantly the ablation targets in sinus rhythm. Of utmost importance, we show that the size of endocardial bipolar DS area is the best predictor of epicardial bipolar DS. Specifically, scar transmurality should be suspected when DS area exceeds 10% area of endocardial surface area, in accordance with MRI measurements of late-enhanced scar (8%). 23

9 Tsiachris et al EAM Features in Post-MI VT Ablation 871 Figure 4. Dense scar is also distributed in septal segments in either anterior or inferolateral infarcts. However, late potentials (LPs) are located in nonseptal segments of the endocardial shell possibly because of the latency of normal activation in the lateral compared with the septal wall. EGM Characteristics in Post-MI Patients Specific areas of LPs are recognized in two third of post-mi patients and complete LP abolition is accomplished in half of the total post-mi population at least endocardially. LPs are expected to be found in post-mi patients with larger and more solid scars based on the increased endocardial bipolar scar density. Such areas of transmural scar, as assessed by echo or computed tomography, have been correlated to areas of LAVAs in post-mi patients. 25 In our study, scar transmurality and density predict reliably the presence of epicardial LPs, indicating the possible requirement for a more complex procedure. The fact that we recognized epicardial LPs as often in research protocol patients as in those with a previous failed endocardial procedure enhances even more the significance of these findings. Based on segmental analysis, two third of LPs are found over segments with DS, whereas only 7% of them in normal myocardium and above unipolar scar (penumbra). However, two third of EPs are located in the border zone and the remaining one third in DS region. We also exhibited a more frequent localization of LPs in nonseptal segments of the endocardial shell (anterior segments in anterior MIs and inferolateral segments in inferolateral MIs), possibly attributed to the latency of normal activation in the lateral wall, as well as to the greater proportion of transmural scar in these segments. 26 We also identified that 80% of LPs in the epicardium are isolated LPs, whereas isolated and fractionated LPs are evenly distributed in the endocardium. Recent findings identified isolated LPs in sites of heterogeneous islets within DS. 27 EPs are found mainly in the endocardium reflecting a possible bridge along channels leading to latest activity and their lower prevalence at epicardial sites might indicate their intramyocardial distribution. Subsequently epicardial isolated LPs could just reflect the resulting conduction pattern. From a pathophysiologic point of view, surviving cells surrounded by fibrosis in areas of inhomogeneous scars are poorly coupled to the rest of the myocardium and constitute the substrate of LPs and EPs. It has recently been exhibited that the presence of LPs rather than EPs increases the specificity for identifying the clinical VT isthmus. 28 Appropriateness of Each Ablation Strategy in Post- MI Patients VT and LPs ablation was the principal ablation strategy in 48%, VT mapping alone in 27%, LPs ablation alone in 17%, and ablation of EPs was the basic strategy in only 6% of the post-mi patients, despite the fact that EPs were recognized in a significantly larger LV area (22.1% versus 10% of endocardial segments) compared with LPs. Although not all areas of LPs contain critical isthmuses for VT maintenance, LPs and selected EPs as defined in the present study provide an ideal substrate for VT ablation because such abnormal EGMs are found more frequently in post-mi patients with spontaneous VT compared with those without arrhythmias. 29 One may hypothesize that

10 872 Circ Arrhythm Electrophysiol August 2015 these sites of slow conduction, that are initially incapable of supporting VT, will form critical isthmuses in the future, representing the necessary if not sufficient arrhythmogenic substrate for re-entrant VT. Of course, this generalized approach toward VT ablation extends the required ablation zone more than specific critical isthmus and channels ablation but less than complete LAVAs elimination and endo epi scar homogenization. Based on the fact that our mean LVA area was >50cm2 (22% of total endocardium), achievement of scar homogenization, beyond its subjective interpretation, necessitates an extended damage in the myocardium, compared with the half of ablated area observed in our study. Targeting only the entrance of conducting channels, that is, the earliest of LPs or encircling and electrically isolating the scar are attractive alternatives, since they minimize the required ablation lesions, but their value has to be proven in larger populations Moreover, electric scar isolation is unlikely to be universally feasible with current ablation technology. We consider that LPs abolition should constitute the primary target for substrate ablation because LPs abolition is a clear end-point amenable to objective evaluation and LPs are more closely related to clinical VT isthmuses. 28 Notably, LPs ablation adds to most cases of VT mapping and constitutes the only option in one quarter of post-mi patients. In contrast, LAVAs elimination is liable to a more subjective interpretation as with current EAM systems it is impossible to provide a complete annotation of presence and distribution of LAVAs before and after ablation. EPs elimination should adjunctively be performed after a definite electrophysiological proof of their involvement in the VT circuit, in patients without LPs (especially those with septal scars) and in cases of persistent VT inducibility after LPs abolition. The development of an improved multipolar recording technology that can identify slow conduction entrances into the scar and allow observing dynamic changes in the slow conducting channels during ablation could allow reliable and prompt EP characterization and subsequent documentation of their disappearance post ablation. Prognostic Value of EAM Characteristics DS extension, with a cutoff value of 25 cm 2, has been proven to be a potent predictor of VT recurrences in post- MI patients. 8 Such an association was not observed in our study population, where LP abolition was associated with reduced VT recurrence. Last but not least, we had the opportunity to assess for the first time the prognostic value of EAM characteristics for cardiac death. Specifically, an increased unipolar scar exceeding bipolar LVA (penumbra) reflects diffuse disorganized myocyte loss and is associated with increased mortality, whereas endocardial LP presence improves prognosis possibly through a successful VT ablation procedure. Limitations This is a single-center prospective analysis of our current ablation strategy not a randomized controlled trial assessing one strategy versus another. Moreover, epicardial maps were less than the endocardial maps and by virtue of this sampling error, there could be a systematic selection bias in the results and a decreased power in the specific analysis. In addition, we have not used a competing risks model for VT recurrence and we acknowledge that there may be confounding variables because we have not performed a statistical model building. The prospective data collection in consecutive patients and the large sample size of patients with detailed EAM analysis constitute the strengths of this study. Conclusions Although DS is not always identified in post-mi patients, its endocardial extension and density predict not only scar transmurality but also the presence of LPs either in the endocardium or in the epicardium. Endocardial LPs, more often present in nonseptal segments and their abolition predicted VT recurrence predominantly in patients with inferolateral MIs, while the size of unipolar penumbra area was associated with cardiac mortality. Acknowledgments We would like to thank Claudio Albertini and Sebastiano Colombo for their technical support and all the staff at the San Raffaele VTU and ICU for their tireless work and professionalism. Disclosures Dr Bella is a consultant for St. Jude Medical and has received honoraria for lectures from Biosense Webster, St Jude Medical and Biotronik. Drs Silberbauer and Oloriz are Advanced European Heart Rhythm Association Fellows with grants funded by Biosense Webster. References 1. Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS, Daubert JP, Higgins SL, Brown MW, Andrews ML; Multicenter Automatic Defibrillator Implantation Trial II Investigators. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med. 2002;346: doi: / NEJMoa Irvine J, Dorian P, Baker B, O Brien BJ, Roberts R, Gent M, Newman D, Connolly SJ. Quality of life in the Canadian Implantable Defibrillator Study (CIDS). Am Heart J. 2002;144: Poole JE, Johnson GW, Hellkamp AS, Anderson J, Callans DJ, Raitt MH, Reddy RK, Marchlinski FE, Yee R, Guarnieri T, Talajic M, Wilber DJ, Fishbein DP, Packer DL, Mark DB, Lee KL, Bardy GH. Prognostic importance of defibrillator shocks in patients with heart failure. N Engl J Med. 2008;359: doi: /NEJMoa Kuck KH, Schaumann A, Eckardt L, Willems S, Ventura R, Delacrétaz E, Pitschner HF, Kautzner J, Schumacher B, Hansen PS; VTACH study group. Catheter ablation of stable ventricular tachycardia before defibrillator implantation in patients with coronary heart disease (VTACH): a multicentre randomised controlled trial. Lancet. 2010;375: doi: / S (09) Carbucicchio C, Santamaria M, Trevisi N, Maccabelli G, Giraldi F, Fassini G, Riva S, Moltrasio M, Cireddu M, Veglia F, Della Bella P. Catheter ablation for the treatment of electrical storm in patients with implantable cardioverter-defibrillators: short- and long-term outcomes in a prospective single-center study. Circulation. 2008;117: doi: / CIRCULATIONAHA 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: Soejima K, Suzuki M, Maisel WH, Brunckhorst CB, Delacretaz E, Blier L, Tung S, Khan H, Stevenson WG. Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation. 2001;104:

11 Tsiachris et al EAM Features in Post-MI VT Ablation Arenal Á, Hernández J, Calvo D, Ceballos C, Atéa L, Datino T, Atienza F, González-Torrecilla E, Eídelman G, Miracle Á, Avila P, Bermejo J, Fernández-Avilés F. Safety, long-term results, and predictors of recurrence after complete endocardial ventricular tachycardia substrate ablation in patients with previous myocardial infarction. Am J Cardiol. 2013;111: doi: /j.amjcard Vergara P, Trevisi N, Ricco A, Petracca F, Baratto F, Cireddu M, Bisceglia C, Maccabelli G, Della Bella P. Late potentials abolition as an additional technique for reduction of arrhythmia recurrence in scar related ventricular tachycardia ablation. J Cardiovasc Electrophysiol. 2012;23: doi: /j x. 10. Jaïs P, Maury P, Khairy P, Sacher F, Nault I, Komatsu Y, Hocini M, Forclaz A, Jadidi AS, Weerasooryia R, Shah A, Derval N, Cochet H, Knecht S, Miyazaki S, Linton N, Rivard L, Wright M, Wilton SB, Scherr D, Pascale P, Roten L, Pederson M, Bordachar P, Laurent F, Kim SJ, Ritter P, Clementy J, Haïssaguerre M. Elimination of local abnormal ventricular activities: a new end point for substrate modification in patients with scar-related ventricular tachycardia. Circulation. 2012;125: doi: /CIRCULATIONAHA Di Biase L, Santangeli P, Burkhardt DJ, Bai R, Mohanty P, Carbucicchio C, Dello Russo A, Casella M, Mohanty S, Pump A, Hongo R, Beheiry S, Pelargonio G, Santarelli P, Zucchetti M, Horton R, Sanchez JE, Elayi CS, Lakkireddy D, Tondo C, Natale A. Endo-epicardial homogenization of the scar versus limited substrate ablation for the treatment of electrical storms in patients with ischemic cardiomyopathy. J Am Coll Cardiol. 2012;60: doi: /j.jacc Tilz RR, Makimoto H, Lin T, Rillig A, Deiss S, Wissner E, Mathew S, Metzner A, Rausch P, Kuck KH, Ouyang F. Electrical isolation of a substrate after myocardial infarction: a novel ablation strategy for unmappable ventricular tachycardias feasibility and clinical outcome. Europace. 2014;16: doi: /europace/eut Della Bella P, Baratto F, Tsiachris D, Trevisi N, Vergara P, Bisceglia C, Petracca F, Carbucicchio C, Benussi S, Maisano F, Alfieri O, Pappalardo F, Zangrillo A, Maccabelli G. Management of ventricular tachycardia in the setting of a dedicated unit for the treatment of complex ventricular arrhythmias: long-term outcome after ablation. Circulation. 2013;127: doi: /CIRCULATIONAHA Hutchinson MD, Gerstenfeld EP, Desjardins B, Bala R, Riley MP, Garcia FC, Dixit S, Lin D, Tzou WS, Cooper JM, Verdino RJ, Callans DJ, Marchlinski FE. Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Circ Arrhythm Electrophysiol. 2011;4: doi: /CIRCEP Chopra N, Tokuda M, Ng J, Reichlin T, Nof E, John RM, Tedrow UB, Stevenson WG. Relation of the unipolar low-voltage penumbra surrounding the endocardial low-voltage scar to ventricular tachycardia circuit sites and ablation outcomes in ischemic cardiomyopathy. J Cardiovasc Electrophysiol. 2014;25: doi: /jce Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS; American Heart Association Writing Group on Myocardial Segmentation and Registration for Cardiac Imaging. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation. 2002;105: Silberbauer J, Oloriz T, Maccabelli G, Tsiachris D, Baratto F, Vergara P, Mizuno H, Bisceglia C, Marzi A, Sora N, Guarracini F, Radinovic A, Cireddu M, Sala S, Gulletta S, Paglino G, Mazzone P, Trevisi N, Della Bella P. Noninducibility and late potential abolition: a novel combined prognostic procedural end point for catheter ablation of postinfarction ventricular tachycardia. Circ Arrhythm Electrophysiol. 2014;7: doi: /CIRCEP Berruezo A, Fernández-Armenta J, Mont L, Zeljko H, Andreu D, Herczku C, Boussy T, Tolosana JM, Arbelo E, Brugada J. Combined endocardial and epicardial catheter ablation in arrhythmogenic right ventricular dysplasia incorporating scar dechanneling technique. Circ Arrhythm Electrophysiol. 2012;5: doi: /CIRCEP Tung R, Mathuria NS, Nagel R, Mandapati R, Buch EF, Bradfield JS, Vaseghi M, Boyle NG, Shivkumar K. Impact of local ablation on interconnected channels within ventricular scar: mechanistic implications for substrate modification. Circ Arrhythm Electrophysiol. 2013;6: doi: /CIRCEP Tilz RR, Makimoto H, Lin T, Rillig A, Deiss S, Wissner E, Mathew S, Metzner A, Rausch P, Kuck KH, Ouyang F. Electrical isolation of a substrate after myocardial infarction: a novel ablation strategy for unmappable ventricular tachycardias feasibility and clinical outcome. Europace. 2014;16: doi: /europace/eut Desjardins B, Crawford T, Good E, Oral H, Chugh A, Pelosi F, Morady F, Bogun F. Infarct architecture and characteristics on delayed enhanced magnetic resonance imaging and electroanatomic mapping in patients with postinfarction ventricular arrhythmia. Heart Rhythm. 2009;6: doi: /j.hrthm Codreanu A, Odille F, Aliot E, Marie PY, Magnin-Poull I, Andronache M, Mandry D, Djaballah W, Régent D, Felblinger J, de Chillou C. Electroanatomic characterization of post-infarct scars comparison with 3-dimensional myocardial scar reconstruction based on magnetic resonance imaging. J Am Coll Cardiol. 2008;52: doi: /j. jacc Wijnmaalen AP, van der Geest RJ, van Huls van Taxis CF, Siebelink HM, Kroft LJ, Bax JJ, Reiber JH, Schalij MJ, Zeppenfeld K. Head-tohead comparison of contrast-enhanced magnetic resonance imaging and electroanatomical voltage mapping to assess post-infarct scar characteristics in patients with ventricular tachycardias: real-time image integration and reversed registration. Eur Heart J. 2011;32: doi: / eurheartj/ehq Wijnmaalen AP, Schalij MJ, von der Thüsen JH, Klautz RJ, Zeppenfeld K. Early reperfusion during acute myocardial infarction affects ventricular tachycardia characteristics and the chronic electroanatomic and histological substrate. Circulation. 2010;121: doi: / CIRCULATIONAHA Komatsu Y, Cochet H, Jadidi A, Sacher F, Shah A, Derval N, Scherr D, Pascale P, Roten L, Denis A, Ramoul K, Miyazaki S, Daly M, Riffaud M, Sermesant M, Relan J, Ayache N, Kim S, Montaudon M, Laurent F, Hocini M, Haïssaguerre M, Jaïs P. Regional myocardial wall thinning at multidetector computed tomography correlates to arrhythmogenic substrate in postinfarction ventricular tachycardia: assessment of structural and electrical substrate. Circ Arrhythm Electrophysiol. 2013;6: doi: /CIRCEP Komatsu Y, Daly M, Sacher F, Derval N, Pascale P, Roten L, Scherr D, Jadidi A, Ramoul K, Denis A, Jesel L, Zellerhoff S, Lim HS, Shah A, Cochet H, Hocini M, Haïssaguerre M, Jaïs P. Electrophysiologic characterization of local abnormal ventricular activities in postinfarction ventricular tachycardia with respect to their anatomic location. Heart Rhythm. 2013;10: doi: /j.hrthm Nakahara S, Tung R, Ramirez RJ, Gima J, Wiener I, Mahajan A, Boyle NG, Shivkumar K. Distribution of late potentials within infarct scars assessed by ultra high-density mapping. Heart Rhythm. 2010;7: doi: /j.hrthm Mountantonakis SE, Park RE, Frankel DS, Hutchinson MD, Dixit S, Cooper J, Callans D, Marchlinski FE, Gerstenfeld EP. Relationship between voltage map channels and the location of critical isthmus sites in patients with post-infarction cardiomyopathy and ventricular tachycardia. J Am Coll Cardiol. 2013;61: doi: /j.jacc Haqqani HM, Kalman JM, Roberts-Thomson KC, Balasubramaniam RN, Rosso R, Snowdon RL, Sparks PB, Vohra JK, Morton JB. Fundamental differences in electrophysiologic and electroanatomic substrate between ischemic cardiomyopathy patients with and without clinical ventricular tachycardia. J Am Coll Cardiol. 2009;54: doi: /j. jacc