Purpose: The purpose of this study was to describe the safety and efficacy of hybrid recanalization procedures in a series of patients with obstructed central veins requiring cardiac implantable electronic device (CIED) revision. Methods: Between 2008 and 2016, 38 consecutive patients (24 M; age 60.5 +/- 16.2 years; range 25-87 years) with central venous obstruction underwent 42 recanalization interventions performed in conjunction with CIED revision or extraction. Fifty percent of patients (19/38) presented with veno-occlusive symptoms, and 13% (5/ 38) of patients had CIED leads with an ipsilateral upper extremity dialysis conduit. Results: Ninety-one percent (38/42) of all procedures resulted in successful recanalization and CIED revision. Twenty-four percent (9/38) of all patients required secondary procedures due to recurrent stenosis, and 78% (7/ 9) of those requiring secondary procedures had indwelling dialysis conduits and/or clinical symptoms related to venous occlusion before the initial procedure. There were complications in 2 patients related to recanalization, and in 3 related to CIED revision. Conclusions: Recanalization of central venous stenosis/occlusion in patients with CIED can be technically challenging but is successful in most patients. Symptomatic patients and those with dialysis conduits often require more aggressive revascularization interventions and may be at increased risk of complication or need for secondary interventions.

Mapping catheter entrapment in a mechanical mitral valve (MV) is a feared but seemingly rare complication.1,2 Here, we describe entrapment, shearing-off, and retrieval of a bipole on the PentaRay mapping (PRM) catheter (Biosense Webster, Inc, Diamond Bar, CA) in a mechanical MV.

BackgroundCardiac resynchronization defibrillator (CRT-D) devices improve survival for New York Heart Association classes II-IV systolic heart failure patients with QRS > 120 ms and left ventricular ejection fraction < 35%. A limitation of 100% CRT pacing is excess battery depletion and pulse generator (PG) replacement compared to VVI or dual-chamber systems. Ampere hour (Ah) measures PG battery capacity and may predict CRT-D device longevity.MethodsWe performed a multicenter retrospective study of all CRT-D devices implanted at our centers from August 1, 2008 to December 31, 2010. Analysis was performed for survival to elective replacement indicator (ERI) between 1.0 Ah, 1.4 Ah, and 2.0 Ah devices, per manufacturers' specifications.ResultsOne thousand three hundred and two patients were studied through December 31, 2014. Patients were followed for an average of 3.0 1.3 years (794 1.0 Ah, 322 2.0 Ah, and 186 1.4 Ah devices under study). CRT-D generator ERI occurred in 13.5% of 1.0 Ah systems (107 out of 794), versus 3.8% in 1.4 Ah (seven out of 186), and 0.3% in 2.0 Ah devices (one out of 322) over mean follow-up of 3.0 years. Odds ratio (OR) for reaching ERI with 1.0 Ah device versus 1.4 Ah or 2.0 Ah was 9.73, P < 0.0001. Univariate predictors for ERI included 1.0 Ah device and LV pacing output >3V @ 1 ms (OR: 3.74, P < 0.001). LV impedance >1,000 ohms predicted improved device survival (OR: 0.38, P = 0.0025).ConclusionsCRT-D battery capacity measured by Ah is a strong predictor of survival to ERI for modern systems. Further study on cost and morbidity associated with early PG change in 1.0 Ah systems is warranted.

Long‐Term Evaluation of Biotronik Linox Family of ICD LeadsIntroductionExpert consensus holds that post‐market, systematic surveillance of ICD leads is essential to ensure confirmation of adequate lead performance. GALAXY (NCT00836589) and CELESTIAL (NCT00810264) are ongoing multicenter, prospective, non‐randomized registries conducted to confirm the long‐term safety and reliability of Biotronik leads.Methods and ResultsICD and CRT‐D patients are followed for Linox and Linoxsmart ICD lead performance and safety for 5 years post‐implant. All procedural and system‐related adverse events (AEs) were assessed at each follow‐up, along with lead electrical parameters. An independent CEC of EPs adjudicated AEs to determine AE category and lead relatedness. The analysis used categories of lead observations per ISO 5841‐2 (Third edition).A total of 3,933 leads were implanted in 3,840 patients (73.0% male, mean age 67.0 ± 12.2 years) at 146 US centers. The estimated cumulative survival probability was 96.3% at 5 years after implant for Linox leads and 96.6% at 4 years after implant for Linoxsmart leads. A comparison of the Linox and Linoxsmart survival functions did not find evidence of a difference (P = 0.2155). The most common AEs were oversensing (23, 0.58%), conductor fracture (14, 0.36%), failure to capture (13, 0.33%), lead dislodgement (12, 0.31%), insulation breach (10, 0.25%), and abnormal pacing impedance (8, 0.20%).ConclusionsLinox and Linoxsmart ICD leads are safe, reliable and infrequently associated with lead‐related AEs. Additionally, estimated cumulative survival probability is clinically acceptable and well within industry standards. Ongoing data collection will confirm the longer‐term safety and performance of the Linox family of ICD leads.

BACKGROUND The natural history of premature ventricular complex (PVC)-induced cardiomyopathy is incompletely understood.OBJECTIVE The purpose of this study was to assess long term follow-up data in patients who underwent successful PVC ablation for PVC-induced cardiomyopathy.METHODS The subjects of this study were 60 patients (17 women; mean age 52.5 +/- 16.8 years; ejection fraction [EF] 37.3 +/- 8.5%, median 40%, interquartile range [IQR] 15) with PVC-induced cardiomyopathy who underwent successful ablation of their predominant PVCs between 2005 and 2012. Patients were followed up for a mean of 23.6 +/- 17.2 months. EF improved to 57.2 +/- 4.7% (median 55%, IQR 5; P = .0001) within 9.6 +/- 8.4 months of the ablation procedure. During follow-up, 10 of 60 patients (16.7%) had recurrent frequent PVCs and 50 patients (83.3%) did not. Patients underwent repeat assessment of EF and PVC burden.RESULTS During follow-up of 23.6 17.2 months, 10 patients had recurrent frequent PVCs, with an increase of their PVC burden from 1.4 +/- 0.9% (median 1.05%, IQR 1.59) after the initial ablation to 27.2 +/- 8.8% (median 26.0%, IQR 18.2; P = .018). Their EF decreased from 55.7 +/- 3.4% (median 55%, IQR 5.8) after the initial ablation to 40.2 +/- 5.1% (median 40%, IQR 15; P = .005). In the remaining patients with PVC-induced cardiomyopathy, EF and PVC burden remained unchanged during follow-up. Patients with PVC recurrence had a higher number of pleomorphic PVC morphologies during initial presentation (4.7 +/- 2.2 vs 2.5 +/- 2.8, P = .002).CONCLUSION Recurrence of frequent PVCs in patients with a history of PVC cardiomyopathy can result in recurrence of cardiomyopathy. Follow-up in patients with PVC-induced cardiomyopathy is important, especially if patients were asymptomatic from the PVCs and have pleomorphic PVCs.

This report describes the percutaneous extraction of embolized intracardiac inferior vena cava (IVC) filter struts using fluoroscopy and fused intracardiac echocardiography and three-dimensional,electroanatomic mapping. Six patients with; indwelling IVC filters placed at outside institutions 5 months to 14 years previously presented with cross-sectional imaging of the chest demonstrating fractured IVC filter struts embolized to the myocardial free wall (four patients) or interventricular septum (two patients). All embolized filter struts were successfully retrieved, and open heart surgery was avoided.

Recurrent atrial fibrillation (AF) after successful cardioversion can be predicted by obstructive sleep apnea (OSA) diagnosed by polysomnography. However, it is not known whether the validated STOP-BANG questionnaire can predict AF recurrence after radiofrequency ablation (RFA). Our objective is to determine the prevalence of unrecognized OSA in patients with AF and its relation to freedom from AF after RFA.

BACKGROUND Frequent idiopathic premature ventricular complexes (PVCs) can result in PVC-induced cardiomyopathy. Frequent PVCs can also aggravate ischemic cardiomyopathy.OBJECTIVE The purpose of this study was to investigate the impact of frequent PVCs on nonischemic cardiomyopathy.METHODS This was a consecutive series of 30 patients (mean age 59.1 +/- 12.1; 18 men; mean ejection fraction [EF] 38% +/- 15%) with structurally abnormal hearts based on the presence of scar on cardiac magnetic resonance imaging and/or a history of cardiomyopathy before the presence of frequent PVCs who were referred for ablation of frequent PVCs.RESULTS Ablation was successful in 18 of 30 patients (60%), resulting in an increase of mean EF from 33.9% +/- 14.5% to 45.7% 17% (P < .0001) during mean follow-up of 30 28 months. The PVC burden in these patients was reduced from 23.1% +/- 8.8% to 1.0% 0.9% (P < .0001). Mean EF did not change in patients with a failed ablation procedure (44.4 +/- 16 vs 43.5 +/- 21, P = .85). The PVC site of origin was in scar tissue in 14 of 18 patients with a successful ablation procedure. Mean New York Heart Association functional class improved from 2.3 +/- 0.6 to 1.1 +/- 0.2 (P < .0001) in patients with a successful outcome and remained unchanged in patients with an unsuccessful outcome (1.9 +/- 0.9 vs 1.9 +/- 0.7, P = 1).CONCLUSION In patients with frequent PVCs and nonischemic cardiomyopathy, EF and functional class can be improved but not always normalized by successful PVC ablation. In most patients with an effective ablation, the arrhythmogenic substrate was located in scar tissue.

BACKGROUND:Ventricular tachycardia (VT) in patients with cardiomyopathy originates in scar tissue. Intramural or epicardial scar may result in ineffective ablation if mapping and ablation are limited to the endocardium. The purpose of this study was to investigate whether preprocedural magnetic resonance imaging (MRI) is beneficial in patients with failed endocardial VT ablations in determining an appropriate ablation strategy. METHODS AND RESULTS:A cardiac MRI was performed in 20 patients with a failed ablation procedure and cardiomyopathy (nonischemic n = 12, ischemic n = 8). A subsequent ablation strategy was determined by a delayed enhanced MRI (DE-MRI) and an epicardial subxyphoid access was planned only in patients with epicardial or intramural free-wall scar. MRIs were performed in all patients with or without an implanted cardioverter defibrillator (ICD). The location of scar tissue in the MRI predicted the origin of VT in all patients. In 9/20 patients an epicardial procedure was performed based on the result of the MRI. An endocardial procedure was performed in the remaining 11 patients who had either endocardial or septal scarring and one patient in whom the MRI only showed artifact. Five patients remained inducible postablation and four patients had VT recurrence within a follow-up period of 17 ± 22 months. All of the latter patients had an intramural scar pattern. CONCLUSIONS:Imaging with DE-MRI prior to VT ablation in patients with previously failed endocardial ablation procedures is beneficial in identifying an ablation strategy, helps to focus on an area of interest intraprocedurally, and provides valuable outcomes information.

To describe techniques used in percutaneous extraction of fractured IVC filter fragments embedded in myocardium. Four patients with indwelling IVC filters presented with cross-sectional imaging demonstrating fractured IVC filter fragments embolized to the free wall (2) or intraventricular septum (2). Two patients complained of chest pain, one of abdominal pain, and one of back pain. Medical records, filter types, devices, techniques, and complications were reviewed. All four patients underwent intracardiac echocardiography (ICE), electroanatomic (EA) mapping, and fluoroscopic guidance to aid in retrieval of the fracture fragments. Filter types included Bard G2 (2), the Bard G1 (1), and Simon Nitinol (1). All four patients had cross sectional imaging of the chest and abdomen. The first patient had a failed transjugular retrieval attempt using standard techniques, and then a second successful attempt using ICE and EA mapping. ICE and EA mapping were used for the remaining 3 patients. Filter struts were captured by a microsnare in 3 patients and vascular retrieval forceps in the 4th patient. In 3 patients, the fracture fragment was retrieved from the right ventricle using the right IJ approach, and in 1 from the left ventricle using a transfemoral arterial route with brief induction of adenosine arrest. In 1 patient, a 0.018” buddy wire was used to penetrate the ventricular septum to stabilize the retrieval sheath in the right ventricle. All 4 fragments were retracted into the ventricular sheath before crossing the tricuspid or aortic valve. The IVC filters and additional fracture fragments in the IVC were removed with standard techniques. There were no complications. With the use of fluoroscopy, ICE, and EA mapping, percutaneous extraction of IVC filter fragments embedded in myocardium is a feasible alternative to open heart surgery.

Introduction: High-output pacing has been advocated as a strategy to avoid injury to the phrenic nerve (PN) during antral pulmonary vein (PV) isolation. We assessed the hypothesis that pacing does not prevent PN injury in patients undergoing radiofrequency (RF) ablation of atrial fibrillation (AF). Methods: The medical records of 198 consecutive patients (age=63±12 years, 129 men, ejection fraction=57±10%, LA=44±6mm, paroxysmal=49%) undergoing their first ablation procedure for AF were reviewed. All patients underwent antral PV isolation using a 3D mapping system (CARTO XP or CARTO 3) and a 3.5 mm irrigated-tip ablation catheter (maximum power, 25 W). Prior to RF energy delivery, high-output pacing (20 mA @ 10 ms, maximum output) was performed to asses for PN capture. Sites that afforded PN capture were avoided and RF energy was delivered at adjacent sites without PN capture. The 3-D maps were reviewed to identify the prevalence and sites of PN capture. Results: High-output pacing along the anterior right antrum resulted in PN capture in 35 patients (18%). The most common site with a positive response was the crux between the upper and lower PVs (60%), followed by the right superior PV (43%), and the right inferior PV (20%). Of the patients with PN capture, 49% had only one site of capture, 20% with two sites, and 31% had 3 or more sites. All PVs were isolated at the end of the procedure. Two patients (1%) developed PN injury (symptom onset on the day after the procedure), which was confirmed on radiography. In neither case was there evidence of PN capture during the procedure. Symptoms resolved in both patients within 3 months, with normalization of radiographic findings. Conclusions: High-output pacing along the anterior right PV antrum yields PN capture in roughly one-fifth of the patients undergoing PV isolation. Despite a negative response to pacing and alteration of the lesion set, PN injury may occur. The reason for this discordance is unknown, but may include the possibility that the capture threshold of the PN exceeds the maximum output of the stimulator, or that RF energy may injure the pericardiophrenic artery, which accompanies the PN. Avoiding high-power or long-duration lesions and high contact force in this region may minimize the risk of PN injury.

BACKGROUND Atrial fibrillation (AF) is associated with a significant increase in the risk of stroke and mortality. It is unclear whether maintaining sinus rhythm (SR) after radiofrequency ablation (RFA) is associated with an improvement in stroke risk and survival.OBJECTIVE The purpose of this study was to determine whether SR after RFA of AF is associated with an improvement in the risk of cerebrovascular events (CVEs) and mortality during an extended 10-year follow-up.METHODS RFA was performed in 3058 patients (age 58 +/- 10 years) with paroxysmal (n = 1888) or persistent AF (n = 1170). The effects of time-dependent rhythm status on CVEs and cardiac and all-cause mortality were assessed using multivariable Cox models adjusted for baseline and time-dependent variables during 11,347 patient-years of follow-up.RESULTS Independent predictors of a higher arrhythmia burden after RFA were age (estimated beta coefficient [beta] = 0.017 per 10 years, 95% confidence interval [CI] 0.006-0.029, P = .003), left atrial (LA) diameter (beta = 0.044 per 5-mm increase in LA diameter, 95% CI 0.034-0.055, P < .0001), and persistent AF (beta = 0.174, 95% CI 0.147-0.201, P < .0001). CVEs and cardiac and all-cause mortality occurred in 71 (2.3%), 33 (1.1%), and 111 (3.6%), respectively. SR after RFA was associated with a significantly lower risk of cardiac mortality (hazard ratio [HR] 0.41, 95% CI 0.20-0.84, P = .015). There was not a significant reduction in all-cause mortality (HR 0.86, 95% CI 0.58-1.29, P = .48) or CVEs (HR 0.79, 95% CI 0.48-1.29, P = .34) in patients who remained in SR after RFA.CONCLUSION Maintenance of SR after RFA is associated with a reduction in cardiovascular mortality in patients with AF.

CRT-D devices offer survival benefit in select NYHA class II-IV systolic heart failure patients with QRS >120ms and LVEF <35%. A limitation of 100% CRT pacing is excess battery drain, and earlier pulse generator (PG) replacement for CRT systems. CRT-D pulse generator change is associated with significant infection risk, morbidity, and cost. Ampere-hour or Ah measures residual PG battery capacity and may predict device longevity.

Outcomes of Catheter Ablation of PVCBackgroundThe purpose of this study was to assess how well acute procedural outcomes predict the clinical outcome of catheter ablation of premature ventricular complexes (PVCs).MethodsA consecutive series of 50 patients (28 women, age: 51 ± 13 years) with frequent PVCs was referred for PVC ablation. Acute failure was defined as inability to eliminate the predominant PVC or recurrence of the predominant PVC within 12 hours. The PVC burden was reassessed 3 months after the ablation procedure. A successful procedure was defined as reduction of the PVC burden at 3 months by ≥80% of the initial burden.ResultsThe procedure was acutely effective in 37 patients (74%) and at 3 months in 40 patients (80%). The presence or absence of the predominant PVC in the 12 hours postablation had the highest accuracy for outcome at 3 months (accuracy: 90%). From among the 13/50 patients (26%) with evidence of acute failure, 4 had a PVC reduction of ≥80% at 3 months and 10 had a PVC reduction of >50% resulting in symptomatic improvement at 3 months.ConclusionThe presence or absence of the predominant PVC within 12 hours postablation best correlated with the 3‐month‐efficacy data. Recurrence of the predominant PVC shortly after ablation did not indicate a procedural failure and the necessity for a repeat procedure. The majority of these patients had a significant, clinically meaningful reduction in their PVC burden. Acute predictors for procedural outcome at 3 months have a high positive but rather low negative predictive value.

Background— Ventricular arrhythmias have been described to originate from intramural locations. Intramural scar can be assessed by delayed-enhanced MRI, but MRIs cannot be performed on every patient. The objective of this study was to assess the value of voltage mapping to detect MRI-defined intramural scar and to correlate the scar with ventricular arrhythmias. Methods and Results— In 15 consecutive patients (3 women; age 55±16 years; ejection fraction, 49±13%) with structural heart disease, intramural scar was detected by delayed-enhanced MRI. All patients underwent endocardial unipolar and bipolar voltage mapping guided by the registered intramural scar. Scar volume by MRI was 11.7±8 cm 3 with a scar thickness of 4.6±0.7 mm and a preserved endocardial/epicardial rim of 3.3±1.6 and 4.8±2.6 mm, respectively. Endocardial bipolar voltage was 1.6±1.73 mV at the scar, 2.12±2.15 mV in a 1 cm perimeter around the scar, and 2.83±3.39 mV in remote myocardium without scar. The corresponding unipolar voltage was 4.94±3.25, 6.59±3.81, and 8.32±3.39 mV, respectively ( P <0.0001). Using receiver–operator characteristic curves, a unipolar cut-off value of 6.78 mV (area under the curve, 0.78) and a bipolar cut-off value of 1.55 mV (area under the curve, 0.69) best separated endocardial measurements overlying scar as compared with areas not overlying a scar. At least 1 intramural ventricular arrhythmia was eliminated in all but 2 patients in this series. Conclusions— Intramural scar can be detected by unipolar and bipolar voltage, unipolar voltage being more useful. Mapping and ablation of intramural arrhythmias originating from an intramural focus can be accomplished.

BACKGROUND The right ventricular outflow tract (RVOT) is the most common site of origin of ventricular arrhythmias (VAs) in patients with idiopathic VAs. A left bundle branch block, inferior axis morphology arrhythmia is the hallmark of RVOT arrhythmias. VAs from other sites of origin can mimic RVOT VAs, and ablation in the RVOT typically fails for these VAs.OBJECTIVE To analyze reasons for failed ablations of RVOT-like VAs.METHODS Among a consecutive series of 197 patients with an RVOT-like electrocardiographic (ECG) morphology who were referred for ablation, 38 patients (13 men; age 46 +/- 14 years; left ventricular ejection fraction 47% +/- 14%) in whom a prior procedure failed within the RVOT underwent a second ablation procedure. ECG characteristics of the VA were compared to a consecutive series of 50 patients with RVOT VAs.RESULTS The origin of the VA was identified in 95% of the patients. In 28 of 38 (74%) patients, the arrhythmia origin was not in the RVOT. The VA originated from intramural sites (n = 8, 21%), the pulmonary arteries (n = 7, 18%), the aortic cusps (n = 6, 16%), and the epicardium (n = 5, 13%). The origin was within the RVOT in 10 (26%) patients. In 2 (5%) patients, the origin could not be identified despite biventricular, aortic, and epicardial mapping. The VA was eliminated in 34 of 38 (89%) patients with repeat procedures. The ECG features of patients with failed RVOT-like arrhythmias were different from the characteristics of RVOT arrhythmias.CONCLUSIONS In patients in whom ablation of a VA with an RVOT-like appearance fails, mapping of the pulmonary artery, the aortic cusps, the epicardium, the left ventricular outflow tract, and the aortic cusps will help identify the correct site of origin. The 12-lead ECG is helpful in differentiating these VAs from RVOT VAs.

BACKGROUND It is not dear whether dabigatran is as safe and effective as uninterrupted anticoagulation with warfarin during radiofrequency catheter ablation (RFA) of atria L fibrillation (AF).OBJECTIVE To compare the safety and efficacy of dabigatran by using a novel administration protocol and uninterrupted anticoagulation with warfarin for periprocedural anticoagulation in patients undergoing RFA of AF.METHODS In this case-control analysis, 763 consecutive patients (mean age 61 +/- 10 years) underwent RFA of AF using dabigatran (N = 191) or uninterrupted warfarin (N = 572) for periprocedural anticoagulation. In all patients, anticoagulation was started >= 4 weeks before RFA. Dabigatran was herd after the morning dose on the day before the procedure and resumed 4 hours after vascular hemostasis was achieved.RESULTS A transesophageal echocardiogram performed in all patients receiving dabigatran did not demonstrate an intracardiac thrombus. There were no thromboembolic complications in either group. The prevalence of major (4 of 191, 2.1%) and minor (5 of 191, 2.6%) breeding complications in the dabigatran group were similar to those in the warfarin group (12 of 572, 2.1%; P = 1.0 and 19 of 572, 3.3%; P = .8, respectively). Pericardia l. tamponade occurred in 2 of 191 (1%) patients in the dabigatran group and in 7 of 572 (1.2%) patients in the warfarin group (P = 1.0). ALL patients who had a pericardia l. tamponade, including 2 in the dabigatran group, had uneventful recovery after perdicardiocentesis. On multivariate analysis, international. normalized ratio (odds ratio [OR] 4.0; 95% confidence interval. [CI] 1.1-15.0; P = .04), clopidogrel use (OR 4.2; 95% CI 1.5-12.3; P = .01), and CHA(2)DS(2)-VASc score (OR 1.4; 95% CI 1.1-1.8; P = .01) were the independent risk factors of breeding complications only in the warfarin group.CONCLUSIONS When herd for approximately 24 hours before the procedure and resumed 4 hours after vascular hemostasis, dabigatran appears to be as safe and effective as uninterrupted warfarin for periprocedural anticoagulation in patients undergoing RFA of AF.

BACKGROUND Most infarct-related ventricular tachycardias (VTs) have an exit site that can be targeted by endocardial ablation. However, some VT reentry circuits have an exit site that is intramural or epicardial Even these circuits may have an endocardial component that can be endocardially ablated. OBJECTIVE To assess the prevalence of postinfarction VTs with a nonendocardial exit site that can be successfully eliminated by endocardial ablation.METHODS Twenty-eight consecutive patients with postinfarction VT (27 men, age 69 +/- 8 years, ejection fraction 0.25% +/- 0.15%) were referred for VT ablation. A total of 213 VTs were inducible (cycle length 378 +/- 100 ms). Pace mapping was performed throughout the scar, and critical sites were identified for 137 VTs (64.5%). Critical sites identified by entrainment mapping and/or pace mapping were divided into exit and nonexit sites depending on the stimulus-QRS/VT cycle length ratio (S-QRS/VT Cl <= 0.3 vs >0.3).RESULTS Endocardial exit sites (S-QRS/VTCl <= 0.3) were identified for 100 of 137 VTs. Only critical nonexit sites were identified for 37 of 137 (27%) VTs. Nonexit sites were confined to a smaller area within the endocardium (1.81 +/- 1.7 cm(2)) and were located within (2.05 +/- 2.79 cm) than did the VT exit sites. Exit sites had a larger area of matching pace maps (3.86 +/- 1.9 cm(2); P <.01) and were at a closer distance to the border zone (0.93 +/- 1.06 cm; P <.01). A total of 133 of 137 VTs were abated. The success rate was similar for VTs in which exit sites were targeted (n = 90 of 100) and VTs in which on ly nonexit sites were targeted (n = 36 of 37) (P =.83).CONCLUSIONS In about one-third of postinfarction VTs for which critical sites were identified, the exit site was not endocardial Critical nonexit sites that are effective for ablation are often within dense scar at a distance from the border zone and can be missed if only the border zone is targeted.