Long-term outcome of surgical cryoablation for refractory ventricular tachycardia in patients with non-ischemic cardiomyopathy

Long-term outcome of surgical cryoablation for refractory ventricular tachycardia in patients... Abstract Aims Limited data exist on the long-term outcome of patients (pts) with non-ischemic cardiomyopathy (NICM) and ventricular tachycardia (VT) refractory to conventional therapies undergoing surgical ablation (SA). We aimed to investigate the long-term survival and VT recurrence in NICM pts with VT refractory to radiofrequency catheter ablation (RFCA) who underwent SA. Methods and results Consecutive pts with NICM and VT refractory to RFCA who underwent SA were included. VT substrate was characterized in the electrophysiology lab and targeted by RFCA. During SA, previous RFCA lesions/scars were identified and targeted with cryoablation (CA; 3 min/lesion; target −150 °C). Follow-up comprised office visits, ICD interrogations and the social security death index. Twenty consecutive patients with NICM who underwent SA (age 53 ± 16 years, 18 males, LVEF 41 ± 20%; dilated CM = 9, arrhythmogenic right ventricular CM = 3, hypertrophic CM = 2, valvular CM = 4, and mixed CM = 2) were studied. Percutaneous mapping/ablation in the electrophysiology lab was performed in 18 and 2 pts had primary SA. During surgery, 4.9 ± 4.0 CA lesions/pt were delivered to the endocardium (2) and epicardium (11) or both (7). VT-free survival was 72.5% at 1 year and over 43 ± 31 months (mos) (range 1–83mos), there was only one arrhythmia-related death. There was a significant reduction in ICD shocks in the 3-mos preceding SA vs. the entire follow-up period (6.6 ± 4.9 vs. 2.3 ± 4.3 shocks/pt, P = 0.001). Conclusion In select pts with NICM and VT refractory to RFCA, SA guided by pre-operative electrophysiological mapping and ablation may be a therapeutic option. Ventricular tachycardia , Non-ischemic cardiomyopathy , Surgical cryoablation What’s new? Limited data exists on the long-term outcome of patients with non-ischemic cardiomyopathy (NICM) and VT refractory to catheter ablation who undergo surgical cryoablation (SA). We examined outcomes after SA in 20 consecutive patients with NICM and VT At 1 year post-SA, the VT-free survival rate was 72.5% and there was a significant reduction in ICD shocks (6.6 ± 4.9 in the 3 months pre-SA vs. 2.3 ± 4.3 shocks per patient over maximum 83 months post-SA, P = 0.001). There was only one arrhythmia-related death over 43 ± 31 months follow-up. In select pts with NICM and VT refractory to catheter ablation, SA guided by pre-operative electrophysiological mapping and ablation may be a therapeutic option. Introduction Prior to the catheter ablation era, surgical ablation (SA) was used to treat ventricular tachyarrhythmias, particularly in patients with healed myocardial infarction where this technique achieved over 80% success in the long-term management of recurrent of ventricular tachycardia (VT).1,2 With the advent of implantable cardioverter-defibrillators and percutaneous catheter ablation over the past two decades, SA is seldom utilized to treat ventricular arrhythmias. However, in some cases, where the ventricular arrhythmias are refractory or inaccessible to catheter ablation, SA may be considered. While there is long-standing experience with the SA technique in patients with ischemic cardiomyopathy (ICM), there is a paucity of data on the operative strategy for SA and long-term outcomes with this technique in patients with non-ischemic cardiomyopathy (NICM). We previously published our single centre experience on SA in eight consecutive patients with NICM where a priori electroanatomic mapping and electrophysiologic study in the electrophysiology lab was used to help guide the ablation approach in the operating room.3 This study demonstrated the feasibility and medium term efficacy of this strategy for successfully treating refractory VT in patients with NICM. In the present study, we aimed to describe our evolving experience with SA in a cohort of NICM patients with sustained VT refractory to catheter ablation, or in NICM patients who were undergoing cardiac surgery for other indications. We also sought to investigate the long-term outcomes of these patients in terms of both survival and VT recurrence. Methods Patient selection Patients >18 years with NICM and VT who underwent SA at the University of Pennsylvania from 2007 to 15 were included. Patients with pure ICM who underwent SA were excluded. All participants provided written informed consent for both the ablation procedure as well as inclusion in our VT ablation registry which was approved by the University of Pennsylvania Health System’s Institutional Review Board. Electronic medical records were retrospectively reviewed for baseline characteristics, clinical, and demographic data (Table 1). Table 1 Baseline patient characteristics N  20  Age (years)  53 ± 16  Males (#)  18  LVEF (%)  41 ± 20  RVEF (%)  37 ± 5  ICD  20  Disease type     Idiopathic dilated cardiomyopathy  9   Hypertrophic cardiomyopathy  2   Right ventricular cardiomyopathy  3   Valvular cardiomyopathy  4   Mixed ischemic/NICM  2  Cardiac MRI  13   Mid-myocardial LGE  3   Subepicardial LGE  3   Septal LGE  6  Prior ablation (# pts)     Total number procedures  1.7 ± 1.0   ENDO procedures  1.4 ± 0.8   EPI procedures  0.5 ± 0.7  Antiarrhythmic drugs (# pts)     Amiodarone  10   Sotalol  3   Quinidine  2   Mexilitine  2   Class IC  2  Comorbidities     Chronic kidney disease  9   Diabetes  4   Obstructive sleep apnea  5   Hyperlipidemia  2   Chronic obstructive pulmonary disease  1   Hypertension  7   Atrial fibrillation  6   Peripheral vascular disease  2   Sudden cardiac arrest  1   Thyroid disease  1   Heart block  2  Time RFCA to surgery (days prior)  46 ± 104 (median 6 days)  N  20  Age (years)  53 ± 16  Males (#)  18  LVEF (%)  41 ± 20  RVEF (%)  37 ± 5  ICD  20  Disease type     Idiopathic dilated cardiomyopathy  9   Hypertrophic cardiomyopathy  2   Right ventricular cardiomyopathy  3   Valvular cardiomyopathy  4   Mixed ischemic/NICM  2  Cardiac MRI  13   Mid-myocardial LGE  3   Subepicardial LGE  3   Septal LGE  6  Prior ablation (# pts)     Total number procedures  1.7 ± 1.0   ENDO procedures  1.4 ± 0.8   EPI procedures  0.5 ± 0.7  Antiarrhythmic drugs (# pts)     Amiodarone  10   Sotalol  3   Quinidine  2   Mexilitine  2   Class IC  2  Comorbidities     Chronic kidney disease  9   Diabetes  4   Obstructive sleep apnea  5   Hyperlipidemia  2   Chronic obstructive pulmonary disease  1   Hypertension  7   Atrial fibrillation  6   Peripheral vascular disease  2   Sudden cardiac arrest  1   Thyroid disease  1   Heart block  2  Time RFCA to surgery (days prior)  46 ± 104 (median 6 days)  MRI-LGE, magnetic resonance imaging-late gadolinium enhancement; LVEF, left ventricular ejection fraction; RVEF, right ventricular ejection fraction; ENDO, endocardial; EPI, epicardial; ICD, implantable cardioverter defibrillator; RFCA, radiofrequency catheter ablation. Catheter mapping and ablation Patients were brought to the electrophysiology lab in the post-absorptive state and percutaneous femoral access was obtained. A retrograde transaortic approach was used to access the left ventricle (LV). A detailed electroanatomic map of the chamber of interest was created during sinus rhythm or right ventricular (RV) pacing. All mapping was performed with a 3.5 mm, open-irrigation–tip catheter (Navistar Thermocool, Biosense Webster, Diamond Bar, CA). In each case, contact electroanatomic mapping (CARTO, Biosense Webster, Diamond Bar, CA) was performed using a fill and colour threshold  ≤15 mm, to ensure homogeneous sampling and representation of the entire endocardial surface area. Particular care was taken to sample more densely the scar and adjoining region.4,5 Epicardial access for mapping and ablation was typically performed in cases where the 12-lead electrocardiogram, pre-procedural cardiac imaging, and/or assessment of bipolar vs. unipolar voltage abnormalities on the electro-anatomic maps suggested the presence of epicardial substrate. Epicardial access was obtained with the technique originally described by Sosa et al.6 The reference values for identifying low-amplitude endocardial and epicardial bipolar and unipolar electrograms were defined as per previously established criteria.5,7–9 Standard techniques were used during the electrophysiology study, including programmed extrastimulation, entrainment mapping, pacemapping, and activation mapping.10 If the VT was not well tolerated or not reproducibly initiated, detailed characterization of the underlying substrate was performed and a substrate based ablation approach was undertaken. Radiofrequency ablation was performed, extending from the border zone to the dense scar, while transecting critical components of the VT circuit. When these regions were in close proximity to anatomic boundaries (mitral or aortic valve), the lesion sets were extended to incorporate these inexcitable areas. Typical settings during lesion creation were power range of 20–50 W and maximum temperature of 45°C, for a total duration of 60–180 s to achieve an impedance drop of 10–18 Ohms. Study protocol For patients included in the study, the three-dimensional electroanatomic maps were analysed retrospectively offline for anatomic scar distribution. The extent of abnormal endocardial and epicardial bipolar voltage signals was quantified by measuring contiguous areas of abnormal electrograms, using the ‘area calculation’ tool available in the 3D mapping system. Procedural reports were reviewed for intra-procedural data, including VT morphology, VT localization, and acute procedural endpoints. Surgical ablation protocol Patients were taken to the operating room in the post-absorptive state and placed under general anaesthesia. Access to the myocardium was mostly achieved through a midline sternotomy in order to obtain full visualization of the cardiac surface. In certain cases, per the surgeon’s discretion, a less invasive approach via partial lower sternotomy was used. Typically, when performing median sternotomy, myocardial protection was achieved with cold cardioplegia after placing patients on cardio-pulmonary bypass. The relevant surfaces of the heart were carefully inspected for visual and palpable scar as well as previously delivered radiofrequency lesions and key anatomic landmarks (coronary arteries, phrenic nerve, valves; Figure 1). Comparison of the observed scar was made to the pre-acquired electroanatomic map and presumed location of the VT exit based on the 12-lead electrocardiogram when available. In select cases, where an endocardial approach was deemed appropriate, endocardial exposure was accomplished through the aortic valve via an aortotomy above the sinus of Valsalva or through the LV apex [in patients undergoing left ventricular assist device (LVAD) placement]. No additional electrophysiologic mapping was performed during the surgical procedure. Cryothermy was applied to the targets (previous radiofrequency lesions, visible scars, presumptive VT exit sites based on 12-lead morphology) using the Surgifrost Surgical Cryoablation System (Medtronic CryoCath LP, QC, Canada). This is an Argon gas based system consisting of a flexible metal probe with an adjustable insulation sheath that can be moulded to conform to the cardiac contours. Typical cryo applications extended for 3 min (including the thawing phase), achieving a minimum temperature of −150 °C. Lesion delivery was assessed in real time through visual whitening and palpable induration of the myocardial surface. Following cryoablation, any additional surgical intervention (valve repair/replacement, coronary artery by-pass, etc.) was performed as clinically indicated and then the hearts were allowed to rewarm. Patients were acutely recovered in the cardiothoracic surgical intensive care unit and then transferred to the step-down unit. Non-invasive programmed stimulation (NIPS) was performed prior to discharge at the treating physician’s discretion. Anti-arrhythmic drugs were continued at discharge. Figure 1 View largeDownload slide Correlation of data from the electrophysiology lab with surgical cryoablation. Case example of a 51-year old male with a history of NICM (LVEF 35%) who presented with sustained VT and recurrent ICD therapies. Two endocardial and two epicardial radiofrequency ablation procedures were performed targeting a right bundle VT (A-B) with repeated late VT terminations during RF application. The epicardial lesion set was limited by phrenic nerve capture and coronary artery distribution (C) near the VT exit site (star). After the patient experienced recurrent VT with >10 ICD shocks, a surgical approach was undertaken. In the OR, prior epicardial RF lesions were identified and surgical cryoablation was performed from the epicardial LV surface (D). The patient has remained VT free off antiarrhythmic drug therapy over 71 months of follow-up. Figure 1 View largeDownload slide Correlation of data from the electrophysiology lab with surgical cryoablation. Case example of a 51-year old male with a history of NICM (LVEF 35%) who presented with sustained VT and recurrent ICD therapies. Two endocardial and two epicardial radiofrequency ablation procedures were performed targeting a right bundle VT (A-B) with repeated late VT terminations during RF application. The epicardial lesion set was limited by phrenic nerve capture and coronary artery distribution (C) near the VT exit site (star). After the patient experienced recurrent VT with >10 ICD shocks, a surgical approach was undertaken. In the OR, prior epicardial RF lesions were identified and surgical cryoablation was performed from the epicardial LV surface (D). The patient has remained VT free off antiarrhythmic drug therapy over 71 months of follow-up. Follow-up After discharge from the hospital, patients were subsequently followed through routine clinic visits at 6 weeks and then every 6 months. Survival and VT episodes were documented during office visits, remote and in-office ICD interrogations and through telephone calls. Mortality was also assessed through the Social Security Death Index. ICD programming and modification or discontinuation of the antiarrhythmic regimen was left to the discretion of the treating electrophysiologist. To account for multiple ICD therapies or electrical storm, all ICD interventions that occurred within a 24 h period were logged as a single event. The total number of ICD shocks was tabulated separately for comparison. There was no blanking period post-operatively. Statistical analysis Continuous variables are reported as mean ± standard deviations. Categorical data are presented as number of cases or percentages. For comparison of continuous variables between two groups, a paired Student’s t-test was performed for normally distributed variables. The Fisher’s exact test was used to analyse categorical data. A P < 0.05 was considered statistically significant. Kaplan–Meier survival curves were generated for all-cause death, death/transplant, VT recurrence post-SA, and VT recurrence after SA or repeat catheter ablation. Patient follow-up data were censored for either transplant or the time of last follow-up. Statistical analyses were performed on SPSSv21.0 software (SPSS, Chicago, IL). Results Over the study period (2007–15), 608 patients with NICM and/or mixed cardiomyopathy underwent catheter ablation of sustained VT at our institution. Eighteen of these patients required SA due to VT being refractory to ≥1 catheter ablation attempt(s) while two additional patients underwent primary SA (concomitant with aortic valve replacement in one patient and LVAD implantation in the other patient). Thus, the final study cohort comprised 20 patients (3.3% of patients with NICM undergoing VT ablation at our institution during the study period). In 15 patients (75%) SA was a planned procedure and in 5 patients (25%) it was performed on an emergent basis due to cardiac tamponade complicating radiofrequency catheter ablation (RFCA) procedures. The cause for cardiac tamponade was RV laceration during epicardial access in four patients and middle cardiac vein laceration in one. Eleven patients (55%) underwent concomitant planned cardiac surgery (five valve surgeries, two epicardial pacing lead placements, one MAZE surgery, two LVAD implantations, and one single coronary vessel bypass). Identification of substrate and ventricular arrhythmias Results of voltage mapping in the electrophysiology lab are listed in Tables 2 and 3. Consistent with our prior observations in patients with NICM, LV unipolar voltage abnormalities were more extensive than bipolar endocardial abnormalities and epicardial bipolar scar was more prominent than the endocardial bipolar scar. Additionally, the distribution of late gadolinium enhancement (LGE) representing scar as seen on cardiac MRI for each patient can be found in Table 3. A total of 62 VTs were induced in the electrophysiology lab. Forty-four VTs (71%) were localized or targeted from the endocardium and 18 were localized or targeted from the epicardium (29%). The primary reasons for failure of catheter ablation were presence of septal or mid-myocardial substrate in seven (35%), serious complication requiring emergent surgery in five (25%), proximity of target site to phrenic nerve and/or coronary arteries in four (20%), and inability to adequately access epicardial substrate due to the presence of dense pericardial adhesions in three (15%). Table 2 Electroanatomic voltage map characteristics and electrophysiologic study Endocardial bipolar low voltage     Perivalvular  11   Septal  6   Apical  1   None  2  Bipolar scar area (cm2)  28 ± 25  Unipolar voltage abnormalities     Perivalvular  13   Septal  11   Apical  5   None  1  Unipolar scar area (cm2)  88 ± 80  Epicardial bipolar low voltage     Perivalvular  4  Epicardial scar area (cm2)  37 ± 37  VT morphologies (total #/pt)  3.4 ± 2.3  VT morphologies (#/pt at last case)  2.7 ± 1.5  VTs haemodynamically tolerated (#/pt)  0.8 ± 1.5  Endocardial bipolar low voltage     Perivalvular  11   Septal  6   Apical  1   None  2  Bipolar scar area (cm2)  28 ± 25  Unipolar voltage abnormalities     Perivalvular  13   Septal  11   Apical  5   None  1  Unipolar scar area (cm2)  88 ± 80  Epicardial bipolar low voltage     Perivalvular  4  Epicardial scar area (cm2)  37 ± 37  VT morphologies (total #/pt)  3.4 ± 2.3  VT morphologies (#/pt at last case)  2.7 ± 1.5  VTs haemodynamically tolerated (#/pt)  0.8 ± 1.5  VT, ventricular tachycardia. Table 3 Clinical, electrophysiological, and imaging details for patients treated with surgical cryoablation for VT Pt #  Age  Aetiology of NICM  LVEF (%)  # ENDO/EPI RFA procedures  # VTs at last RFA  Clinical VTs at last RFA  Reason for RFCA failure  EP substrate (ENDO)  EP substrate (unipolar)  EP substrate (EPI)  MRI substrate (LGE)  1  52  Valvular  30  1/1  4  VT1: RBRS CL 450 ms  Phrenic nerve location  Perivalvular LV  Septum (base), anterior and anterolateral (base to mid), and lateral (basal)  Anterolateral (base to mid)  Anterior (base to mid) and septum  VT2: bidirectional  VT3: RBRI CL 420 ms  VT4: RBI CL 427 ms  2  72  NICM  50  2/0  1  VT1: RBRI, CL 295  Dense pericardial adhesions  None  Inferoseptal (base)  N/A  N/A  3  55  HOCM  65  2/1  5  VT1: RBRI CL 440  Septal substrate  Lateral  Lateral  Perivalvular (base)  Transmural septum (base to mid), subendocardial lateral (base), inferolateral  VT2: RBRI early transition  VT3: RBLI  VT4: RBI CL 500  VT5: RBI CL 280  4  65  Mixed  10  1/0  3  VT1: LBLI CL 370  Septal substrate  Septum (base)  Septum (base), inferolateral (base), lateral (base), periaortic  N/A  N/A  VT2: RBRS CL 290  VT3: LBLI CL 260  5  58  HOCM  70  1/1  2  VT1: RBS CL 294    LV apex  Septum (base), LV apex, lateral  N/A  Mid-myocardial lateral (base), anterolateral papillary muscle, inferior (mid), anteroseptum  VT2: LBLS CL 234  6  52  NICM  30  2/2  1  VT1: RBLS CL 400  Left circumflex coronary artery and phrenic nerve location  Periaortic  Septum (base)  Lateral  Mid-myocardial to epicardial inferolateral (base), inferior  7  43  Valvular  20  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  N/A  8  75  NICM  35  1/0  1  VT1: RBRI CL 300  Mid-myocardial substrate  Perimitral, periaortic  Diffuse LV (except apex)  N/A  N/A  9  48  NICM  55  3/1  3  VT1: RBRS CL 275  Mid-myocardial substrate  Septum (base), periaortic  Septum (base to mid), perimitral, anterolateral (patchy)  N/A  Mid-myocardial septum (base), anterior (base), inferior (base)  VT2: RBRI CL 325  VT3: RBRI CL 400  10  69  NICM  15  2/0  2  VT1: RBLI CL 450    Periaortic, inferoseptum, inferior (mid, patchy)  Septum (base to apex), inferior, perivalvular.  N/A  N/A  VT2: RBRS  11  51  Valvular  60  1/0  1  VT1: RBLS  Septal substrate  None  None  N/A  Subepicardial inferoseptum (mid)  12  49  ARVC  65  2/0  4  VT1: LBLS CL 243    Inferolateral RV (base to mid)  Inferolateral RV (base to mid)  N/A  Anteroseptal RV (base, patchy)  VT2: LBLS CL 250  VT3: LBLS CL 210  VT4: LBLS CL 194  13  70  Valvular  25  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  Subendocardial inferolateral (base to mid)  14  22  NICM  56  1/1  1  VT1: LBLS  Phrenic nerve location  LV septum (mid, patchy)  None  RV free wall, LV apex  Subepicardial anterior (mid to apex), lateral (apex), RV insertion  15  66  Mixed  20  1/0  3  VT1: RBRI CL 630  Dense pericardial adhesions  Inferoseptum (base to mid), periaortic  Diffuse LV (except lateral wall)  N/A  N/A  VT2: LBRS CL 450  VT3: RBRI CL 445  16  58  NICM  20  1/0  5  VT1: RBRS CL 309    Septum and Inferoseptum (base to mid)  Diffuse LV  N/A  N/A  VT2: RBRS CL 241  VT3: RBS  VT4: RBRS  VT5: RBRS  17  28  ARVC  65  3/2  4  VT1: LBLS (no trans)  LAD and PDA coronary arteries  RV inferior, inferolateral, anterior  Entire RV  N/A  Anterolateral RV, inferior RV  VT2: LBLS (V5 trans)  VT3: LBLS (V5 trans)  VT4: LBLS (V6 trans)  18  59  NICM  25  1/0  2  VT1: RB CL 300  Mid-myocardal substrate  Lateral (base, patchy)  Septum (base), inferior (base to mid), anterolateral perivalvular (base, patchy)  N/A  Subendocardial lateral (base to mid)  VT2: RB CL 288  19  18  ARVC  60  2/0  5  VT1: LBL CL 240  Dense pericardial adhesions  Anterior RVOT free wall and peritricuspid  Diffuse RV  N/A  None  VT2: LBL CL 300  VT3: LBI CL 280  VT4: LBS CL 290  VT5: LBS CL 280  20  58  NICM  40  1/0  1  VT1: RBI CL 300  Mid-myocardial substrate  Periaortic  Periaortic, AMC and anterolateral (base)  N/A  None  Pt #  Age  Aetiology of NICM  LVEF (%)  # ENDO/EPI RFA procedures  # VTs at last RFA  Clinical VTs at last RFA  Reason for RFCA failure  EP substrate (ENDO)  EP substrate (unipolar)  EP substrate (EPI)  MRI substrate (LGE)  1  52  Valvular  30  1/1  4  VT1: RBRS CL 450 ms  Phrenic nerve location  Perivalvular LV  Septum (base), anterior and anterolateral (base to mid), and lateral (basal)  Anterolateral (base to mid)  Anterior (base to mid) and septum  VT2: bidirectional  VT3: RBRI CL 420 ms  VT4: RBI CL 427 ms  2  72  NICM  50  2/0  1  VT1: RBRI, CL 295  Dense pericardial adhesions  None  Inferoseptal (base)  N/A  N/A  3  55  HOCM  65  2/1  5  VT1: RBRI CL 440  Septal substrate  Lateral  Lateral  Perivalvular (base)  Transmural septum (base to mid), subendocardial lateral (base), inferolateral  VT2: RBRI early transition  VT3: RBLI  VT4: RBI CL 500  VT5: RBI CL 280  4  65  Mixed  10  1/0  3  VT1: LBLI CL 370  Septal substrate  Septum (base)  Septum (base), inferolateral (base), lateral (base), periaortic  N/A  N/A  VT2: RBRS CL 290  VT3: LBLI CL 260  5  58  HOCM  70  1/1  2  VT1: RBS CL 294    LV apex  Septum (base), LV apex, lateral  N/A  Mid-myocardial lateral (base), anterolateral papillary muscle, inferior (mid), anteroseptum  VT2: LBLS CL 234  6  52  NICM  30  2/2  1  VT1: RBLS CL 400  Left circumflex coronary artery and phrenic nerve location  Periaortic  Septum (base)  Lateral  Mid-myocardial to epicardial inferolateral (base), inferior  7  43  Valvular  20  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  N/A  8  75  NICM  35  1/0  1  VT1: RBRI CL 300  Mid-myocardial substrate  Perimitral, periaortic  Diffuse LV (except apex)  N/A  N/A  9  48  NICM  55  3/1  3  VT1: RBRS CL 275  Mid-myocardial substrate  Septum (base), periaortic  Septum (base to mid), perimitral, anterolateral (patchy)  N/A  Mid-myocardial septum (base), anterior (base), inferior (base)  VT2: RBRI CL 325  VT3: RBRI CL 400  10  69  NICM  15  2/0  2  VT1: RBLI CL 450    Periaortic, inferoseptum, inferior (mid, patchy)  Septum (base to apex), inferior, perivalvular.  N/A  N/A  VT2: RBRS  11  51  Valvular  60  1/0  1  VT1: RBLS  Septal substrate  None  None  N/A  Subepicardial inferoseptum (mid)  12  49  ARVC  65  2/0  4  VT1: LBLS CL 243    Inferolateral RV (base to mid)  Inferolateral RV (base to mid)  N/A  Anteroseptal RV (base, patchy)  VT2: LBLS CL 250  VT3: LBLS CL 210  VT4: LBLS CL 194  13  70  Valvular  25  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  Subendocardial inferolateral (base to mid)  14  22  NICM  56  1/1  1  VT1: LBLS  Phrenic nerve location  LV septum (mid, patchy)  None  RV free wall, LV apex  Subepicardial anterior (mid to apex), lateral (apex), RV insertion  15  66  Mixed  20  1/0  3  VT1: RBRI CL 630  Dense pericardial adhesions  Inferoseptum (base to mid), periaortic  Diffuse LV (except lateral wall)  N/A  N/A  VT2: LBRS CL 450  VT3: RBRI CL 445  16  58  NICM  20  1/0  5  VT1: RBRS CL 309    Septum and Inferoseptum (base to mid)  Diffuse LV  N/A  N/A  VT2: RBRS CL 241  VT3: RBS  VT4: RBRS  VT5: RBRS  17  28  ARVC  65  3/2  4  VT1: LBLS (no trans)  LAD and PDA coronary arteries  RV inferior, inferolateral, anterior  Entire RV  N/A  Anterolateral RV, inferior RV  VT2: LBLS (V5 trans)  VT3: LBLS (V5 trans)  VT4: LBLS (V6 trans)  18  59  NICM  25  1/0  2  VT1: RB CL 300  Mid-myocardal substrate  Lateral (base, patchy)  Septum (base), inferior (base to mid), anterolateral perivalvular (base, patchy)  N/A  Subendocardial lateral (base to mid)  VT2: RB CL 288  19  18  ARVC  60  2/0  5  VT1: LBL CL 240  Dense pericardial adhesions  Anterior RVOT free wall and peritricuspid  Diffuse RV  N/A  None  VT2: LBL CL 300  VT3: LBI CL 280  VT4: LBS CL 290  VT5: LBS CL 280  20  58  NICM  40  1/0  1  VT1: RBI CL 300  Mid-myocardial substrate  Periaortic  Periaortic, AMC and anterolateral (base)  N/A  None  a Patients 7 and 13 underwent primary SA at the time of LVAD implantation and aortic valve replacement, respectively. Surgical ablation Surgical ablation was performed a median of 7 days after the last catheter ablation (range 0–368 days). Details from the SA procedure for each patient can be found in Table 4. The primary reasons for undergoing surgery were: VT not amenable to or refractory to RFCA in 12 (60%), emergency repair of catheter ablation complication in 5 (25%), LVAD implantation in 2 (10%), and aortic valve replacement in 1 (5%). Surgical ablation was performed via midline sternotomy in 18 patients and via partial lower sternotomy in 2 patients. Seventeen SAs were performed on-pump and three were done off-pump. Cross clamp time was 66 ± 29 min with a bypass time of 120 ± 39 min. An average of 4.9 ± 4.0 (median 3) cryoablation applications were delivered per patient and the mean cryothermy time was 17 ± 14 min. In 2 patients only the endocardium was targeted and in 11 patients only the epicardium was targeted. Seven patients had cryo-lesions delivered to both the endocardium and epicardium. Endocardial access for cryoablation was obtained via ascending aortotomy (seven patients), atriotomy (two patients), or stab puncture through the RV outflow tract (one patient). The phrenic nerve was deflected and coronary arteries were dissected and displaced to allow for targeted cryoablation of the substrate in the four patients who had previously failed catheter ablation for that reason. In all seven patients who had previously failed catheter ablation due to the presence of septal or mid-myocardial substrate, the surgeons were able to target this location and achieve palpable transmurality of lesions with cryoablation. Table 4 Surgical details for patients treated with surgical cryoablation for VT Pt #  Interval between last RFCA and SA (Days)  Main reason for surgery  On/off pump  Pump time (min)  Cross-clamp time (min)  Surgical access  Additional cardiac surgery  ENDO/EPI cryoablation  Endo access  Cryoablation location  1  13  VT, severe MR  On  132  72  MS  MV repair, TV repair, LV epicardial lead  ENDO+EPI  Stab puncture through RVOT  Anterior RV septum (ENDO) and RV lateral to LAD (EPI)  2  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  Evacuation of pericardial hematoma, RV laceration repair  EPI  –  LV basal lateral and basal inferior  3  17  VT, AF  On  189  111  MS  MAZE  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  4  13  Severe MR (MV ring dehiscence) and VT  On  159  110  MS  Redo MV repair, removal of epicardial adhesions, LV epicardial lead  ENDO+EPI  Incision through LA suture line  Posterolateral septum (ENDO); Posterior, inferior, and posterolateral (EPI)  5  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  Inferolateral and inferior portions of LV apex  6  31  VT  On  77  42  MS  None  ENDO+EPI  Ascending aortotomy  LV posterior wall (ENDO+EPI)  7  N/Aa  LVAD implantation, VT  On  152  66  MS  AVR, LVAD  EPI only  –  Epicardium overlying RV/LV  8  7  VT  On  155  34  MS  None  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  9  5  VT  On  110  59  MS  LV epicardial lead  ENDO+EPI  Ascending aortotomy  LV septum and free wall (ENDO); anterior and posterior LV near septum (EPI)  10  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  RV laceration repair, LV epicardial lead  EPI only  –  LV inferior wall  11  31  Severe AR, VT  On  125  96  MS  AVR due to severe AI from LCC perforation  ENDO only  Ascending aortotomy  LV inferoseptum  12  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  RV near acute marginal branch  13  N/Aa  Severe AS, severe MR, VT  On  113  74  MS  AVR, MV repair, MAZE  ENDO only  Ascending aortotomy, left atriotomy at interatrial groove  LV inferior wall  14  289  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  Near RV acute marginal, RV and LV on either side of PDA, LV inferolateral and inferoapex  15  6  VT, CAD  On  81  46  MS  CABG, epicardial LV lead  EPI only  –  LV base, adjacent to CS lead  16  41  Continued VT and need for LVAD implantation  On  62  17  MS  LVAD  EPI only  –  LV inferolateral  17  5  VT  On  N/A  N/A  MS  None  EPI only  –  RV aspect of LAD and RVOT; along basal TV; parallel to RPDA  18  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  MCV tear repair  EPI only  –  LV inferolateral and posterolateral  19  4  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  RVOT to PV; RV anterior free wall to RV inferior wall  20  368  Severe MR and VT  On  87  68  MS  MV repair  ENDO+EPI  Ascending aortotomy  LVOT (ENDO) and anterior to aortic valve (EPI)  Pt #  Interval between last RFCA and SA (Days)  Main reason for surgery  On/off pump  Pump time (min)  Cross-clamp time (min)  Surgical access  Additional cardiac surgery  ENDO/EPI cryoablation  Endo access  Cryoablation location  1  13  VT, severe MR  On  132  72  MS  MV repair, TV repair, LV epicardial lead  ENDO+EPI  Stab puncture through RVOT  Anterior RV septum (ENDO) and RV lateral to LAD (EPI)  2  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  Evacuation of pericardial hematoma, RV laceration repair  EPI  –  LV basal lateral and basal inferior  3  17  VT, AF  On  189  111  MS  MAZE  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  4  13  Severe MR (MV ring dehiscence) and VT  On  159  110  MS  Redo MV repair, removal of epicardial adhesions, LV epicardial lead  ENDO+EPI  Incision through LA suture line  Posterolateral septum (ENDO); Posterior, inferior, and posterolateral (EPI)  5  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  Inferolateral and inferior portions of LV apex  6  31  VT  On  77  42  MS  None  ENDO+EPI  Ascending aortotomy  LV posterior wall (ENDO+EPI)  7  N/Aa  LVAD implantation, VT  On  152  66  MS  AVR, LVAD  EPI only  –  Epicardium overlying RV/LV  8  7  VT  On  155  34  MS  None  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  9  5  VT  On  110  59  MS  LV epicardial lead  ENDO+EPI  Ascending aortotomy  LV septum and free wall (ENDO); anterior and posterior LV near septum (EPI)  10  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  RV laceration repair, LV epicardial lead  EPI only  –  LV inferior wall  11  31  Severe AR, VT  On  125  96  MS  AVR due to severe AI from LCC perforation  ENDO only  Ascending aortotomy  LV inferoseptum  12  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  RV near acute marginal branch  13  N/Aa  Severe AS, severe MR, VT  On  113  74  MS  AVR, MV repair, MAZE  ENDO only  Ascending aortotomy, left atriotomy at interatrial groove  LV inferior wall  14  289  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  Near RV acute marginal, RV and LV on either side of PDA, LV inferolateral and inferoapex  15  6  VT, CAD  On  81  46  MS  CABG, epicardial LV lead  EPI only  –  LV base, adjacent to CS lead  16  41  Continued VT and need for LVAD implantation  On  62  17  MS  LVAD  EPI only  –  LV inferolateral  17  5  VT  On  N/A  N/A  MS  None  EPI only  –  RV aspect of LAD and RVOT; along basal TV; parallel to RPDA  18  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  MCV tear repair  EPI only  –  LV inferolateral and posterolateral  19  4  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  RVOT to PV; RV anterior free wall to RV inferior wall  20  368  Severe MR and VT  On  87  68  MS  MV repair  ENDO+EPI  Ascending aortotomy  LVOT (ENDO) and anterior to aortic valve (EPI)  Follow-up Early/in-hospital follow-up Eighteen of twenty patients (90%) survived to hospital discharge. Cause of death in the two patients who died before discharge was recurrent VT/VF shortly after SA in one patient and sepsis in the other. Mean hospital duration was 28 ± 34 days with a mean time from surgery to discharge of 22 ± 32 days (median 9.5 days). Patients spent a mean of 1.5 ± 1.1 days on mechanical ventilation after the surgery. Non-invasive programmed stimulation was performed in 10 (50%) patients during the index hospitalization. In one patient this was performed in the OR after the heart was rewarmed and in nine patients it was performed pre-discharge at an average of 9 ± 12 days after SA. In 9 of 10 patients (90%), VT was non-inducible at NIPS. Long-term follow-up Follow-up details for each patient are listed in Table 5. Long-term follow-up data revealed a 72.5% VT-free survival at 1-year, 59.3% VT-free survival at 2-years, and 34.6% VT-free survival at 5-years follow-up (Figure 2A). Over a mean follow-up duration of 43 ± 31 months (median 20 months), eight patients died (four pump failure, one refractory VT/VF, two malignancy, one septic shock). Of the 12 living patients, 1 underwent orthotopic heart transplant. Figure 2B shows survival free from death and Figure 2C shows freedom from death or transplant after SA. Of the 11 remaining patients, 5 experienced recurrent VT. Mean time to VT recurrence was 11 ± 15 months. In these five patients freedom from VT was achieved by an antiarrhythmic drug in three, left ganglionic sympathectomy in one, and repeat catheter ablation in four (median 13.8 months after SA). Figure 2D depicts freedom from VT after last surgical/catheter ablation procedure (allowing for repeat catheter ablation attempts after SA). Ten patients were on antiarrhythmic drugs at the last clinic follow-up visit (7 Amiodarone, 1 Amiodarone and Mexiletine, 1 Sotalol). Of the nine patients who were non-inducible for VT at NIPS, five (56%) experienced no subsequent VT recurrences. Table 5 Acute and long-term outcomes after surgical cryoablation for VT Pt #  Age  Year  Aetiology of NICM  VT storm pre-SA  ICD shocks per month in 3 months pre-SA  NIPS post-SA  Inducible at post-SA NIPS  VT storm post- SA  ICD shocks per month long-term post-SA  AAD drug post-SA  Duration of follow-up (months)  Outcome  1  52  2007  Valvular  Yes  2.3  Yes  No  No  0  Beta- blockerQuinidine  3  Deceased (pump failure)  2  72  2008  NICM  No  2.3  Yes  No  No  0  Beta-blocker  78  Alive  3  55  2008  HOCM  No  1  No  –  No  0  Sotalol  76  Alive  4  65  2008  Mixed  Yes  2.7  No  –  No  0  Amiodarone  18  Deceased (pump failure)  5  58  2008  HOCM  No  1  No  –  No  0.03  Amiodarone  72  Repeat catheter ablation  Deceased (malignancy)  6  52  2009  NICM  No  4  Yes  No  No  0  Beta-blocker  71  Alive  7  43  2009  Valvular  Yes  3.7  No  –  No  0.1  Amiodarone  35  Transplant  8  75  2009  NICM  Yes  5.3  No  –  Yes  3.7  Amiodarone  5  Repeat catheter ablation  Sympathectomy  Deceased (sepsis)  9  48  2009  NICM  No  2  Yes  Yes, non-clinical only  No  0.06  Amiodarone  64  Alive  10  69  2010  NICM  No  1.7  No  –  No  0  Amiodarone  2  Deceased (pump failure)  11  51  2010  Valvular  No  –  Yes  No  No  0  Beta-blocker  48  Alive  12  49  2011  ARVC  Yes  0.3  Yes  No  No  0  Beta-blocker  40  Alive  13  70  2011  Valvular  No  0.3  No  –  No  0.2  Beta-blocker  23  Repeat catheter ablation  Deceased  (Pump failure)  14  22  2012  NICM  No  0  Yes  No  No  0.03  Beta-blocker  33  Repeat catheter ablation  Alive  15  66  2012  Mixed  Yes  4.3  No  –  No  2  Beta-blocker  0  Deceased (VT/VF)  16  58  2013  NICM  No  2.3  No  –  No  0  Amiodarone  7  Deceased (malignancy)  17  28  2013  ARVC  No  0.7  Yes  No  No  0  Beta-blocker  14  Alive  18  59  2014  NICM  No  4.7  No  –  No  0.3  Amiodarone  8  Alive  Mexilitine  19  18  2014  ARVC  No  2.7  Yes  No  No  2.8  Beta-blocker  1  Sympathectomy  Alive  20  58  2015  NICM  No  0.3  No  –  No  0  Beta-blocker  1  Alive  Pt #  Age  Year  Aetiology of NICM  VT storm pre-SA  ICD shocks per month in 3 months pre-SA  NIPS post-SA  Inducible at post-SA NIPS  VT storm post- SA  ICD shocks per month long-term post-SA  AAD drug post-SA  Duration of follow-up (months)  Outcome  1  52  2007  Valvular  Yes  2.3  Yes  No  No  0  Beta- blockerQuinidine  3  Deceased (pump failure)  2  72  2008  NICM  No  2.3  Yes  No  No  0  Beta-blocker  78  Alive  3  55  2008  HOCM  No  1  No  –  No  0  Sotalol  76  Alive  4  65  2008  Mixed  Yes  2.7  No  –  No  0  Amiodarone  18  Deceased (pump failure)  5  58  2008  HOCM  No  1  No  –  No  0.03  Amiodarone  72  Repeat catheter ablation  Deceased (malignancy)  6  52  2009  NICM  No  4  Yes  No  No  0  Beta-blocker  71  Alive  7  43  2009  Valvular  Yes  3.7  No  –  No  0.1  Amiodarone  35  Transplant  8  75  2009  NICM  Yes  5.3  No  –  Yes  3.7  Amiodarone  5  Repeat catheter ablation  Sympathectomy  Deceased (sepsis)  9  48  2009  NICM  No  2  Yes  Yes, non-clinical only  No  0.06  Amiodarone  64  Alive  10  69  2010  NICM  No  1.7  No  –  No  0  Amiodarone  2  Deceased (pump failure)  11  51  2010  Valvular  No  –  Yes  No  No  0  Beta-blocker  48  Alive  12  49  2011  ARVC  Yes  0.3  Yes  No  No  0  Beta-blocker  40  Alive  13  70  2011  Valvular  No  0.3  No  –  No  0.2  Beta-blocker  23  Repeat catheter ablation  Deceased  (Pump failure)  14  22  2012  NICM  No  0  Yes  No  No  0.03  Beta-blocker  33  Repeat catheter ablation  Alive  15  66  2012  Mixed  Yes  4.3  No  –  No  2  Beta-blocker  0  Deceased (VT/VF)  16  58  2013  NICM  No  2.3  No  –  No  0  Amiodarone  7  Deceased (malignancy)  17  28  2013  ARVC  No  0.7  Yes  No  No  0  Beta-blocker  14  Alive  18  59  2014  NICM  No  4.7  No  –  No  0.3  Amiodarone  8  Alive  Mexilitine  19  18  2014  ARVC  No  2.7  Yes  No  No  2.8  Beta-blocker  1  Sympathectomy  Alive  20  58  2015  NICM  No  0.3  No  –  No  0  Beta-blocker  1  Alive  Figure 2 View largeDownload slide Kaplan–Meier curves for patients with a NICM who underwent surgical cryoablation for refractory VT at our institution. (A) Survival free of VT after SA. (B) Survival free of death from any cause. (C) Survival free from death or heart transplantation. (D) Survival free of VT (after SA or last catheter ablation). Figure 2 View largeDownload slide Kaplan–Meier curves for patients with a NICM who underwent surgical cryoablation for refractory VT at our institution. (A) Survival free of VT after SA. (B) Survival free of death from any cause. (C) Survival free from death or heart transplantation. (D) Survival free of VT (after SA or last catheter ablation). In terms of total ICD interventions, there was a non-significant trend towards a reduction of total VT events from the 3 months pre-SA compared to the entire post-SA follow-up period (maximum 83 months; 5.7 ± 6.0 vs. 3.2 ± 5.9 events/pt, median 3.0 vs. 1.0 events, P = 0.11; Figure 3). There was a significant reduction in ICD shocks from the 3 months pre-SA to the first 3 months post-SA (mean 6.6 ± 4.9 vs. 1.7 ± 4.3 shocks/pt, respectively; median 7.0 vs. 0.0 events, respectively; P < 0.001) as well as the entire post-SA follow-up duration of 83 months (6.6 ± 4.9 vs. 2.3 ± 4.3 shocks/pt, respectively; median 7.0 vs. 0.0 shocks/pt, respectively; P = 0.001; Figure 3). Of the six patients in our cohort who had experienced VT storm in the interval between their last RFCA attempt and the SA, only one had recurrent VT storm after SA. Figure 3 View largeDownload slide VT events and ICD shocks in patients with NICM undergoing surgical cryoablation for refractory VT. There was a significant reduction in ICD shocks per patient and a non-significant trend towards a reduction in total VT events. Figure 3 View largeDownload slide VT events and ICD shocks in patients with NICM undergoing surgical cryoablation for refractory VT. There was a significant reduction in ICD shocks per patient and a non-significant trend towards a reduction in total VT events. Overall survival based on the aetiology of underlying cardiomyopathy were as follows: arrhythmogenic right ventricular cardiomyopathy (ARVC)—100%, idiopathic dilated cardiomyopathy (IDCM)—67%, hypertrophic cardiomyopathy (HCM)—50%, and valvular cardiomyopathy—25%. These differences, however, were not significant (P = 0.13). There were no significant differences in VT recurrence rates between the subtypes of NICM (ARVC: 33%, IDCM: 44%, valvular cardiomyopathy: 75%, and HCM: 100%; P = 0.32). Discussion While RFCA is an effective treatment option for many patients with VT and NICM, standard techniques for endocardial and epicardial RFCA may be inadequate in a minority of patients. Common causes of failure of percutaneous RFCA for VT include prohibitive proximity of target site to coronary arteries or phrenic nerve, failure to gain percutaneous epicardial access due to adhesions from prior cardiac surgery, and the presence of mid-myocardial or septal substrate. We report our institution’s long-term experience of SA in a consecutive group of NICM patients who suffered from recurrent, symptomatic VT refractory to medical therapy, and percutaneous RFCA. For some of these patients who were not considered transplant candidates, this was a treatment option of last resort. In other patients this was offered during open heart surgery for other indications. In 90% of cases, SA was guided by information gathered in the electrophysiology lab utilizing electroanatomic, activation and pace mapping to characterize the underlying substrate and define critical components of the re-entrant VT circuit. Our long-term outcomes in this critically ill patient population showed a 72.5% freedom from VT at 1 year and a significant reduction in ICD shocks over the long-term. The use of anti-arrhythmic drugs was also reduced from 95 to 50% over the course of the extended follow-up. Overall survival was 60% over 7 years and arrhythmic death was rare. Ablation technique and endpoints Surgical cryoablation for VT in patients with NICM is rarely needed and was used in <5% of our total ablation population. In these unusual cases, we speculate several reasons why SA may have been effective. First, our strategy of utilizing data from the electrophysiology lab for guidance together with surgical access and cold cardioplegia enabled us to more precisely target the VT substrate in the OR under direct visualization of all cardiac surfaces. Secondly, scar is frequently mid-myocardial in patients with NICM. The current RFCA technology may not be able to create deep enough lesions in some patients. This is particularly true when targeting the basal epicardial region where the myocardium is thick and the presence of fat may hinder adequate lesion creation from the epicardium. Delivering cryoablation in the OR under cold cardioplegia may allow for creation of larger lesions.11–13 Technical issues such as catheter stability and contact force during percutaneous catheter ablation are not encountered during surgical cryoablation. Furthermore, structures such as epicardial coronary arteries and the phrenic nerve can be clearly visualized in the OR and manually deflected if ablation is required in their vicinity. Also, dense pericardial adhesions can make percutaneous epicardial access challenging. Ideally, we would have preferred to induce and map the VT in the OR similar to the approach that is used in the electrophysiology lab. However, this is not always feasible for a variety of reasons. First, it requires similar mapping environment and tools as in the electrophysiology lab. Secondly, it can be difficult to induce and sustain VT under cold cardioplegia. The latter scenario also poses a challenge to assessing the intraprocedural success of the ablation procedure. In our series, the surgeons used alternate intra-operative endpoints to assess for adequacy of lesion creation in the OR including visualizing tissue whitening at the surface where the lesion is delivered and feeling for palpable induration on the surface opposing the site of lesion creation (epicardial surface for endocardial lesions and vice-versa). Another criteria that we intend to test in future cases to validate effective lesion delivery in the OR is local inexcitability with high output pacing (10–50 mA × 2 ms). Additional assessment of procedural efficacy may be to perform programmed stimulation after warming the heart. This can be challenging especially if additional cardiac surgery is being contemplated immediately following SA as was the case in more than half of our study population. In the future, defining myocardial voltage changes post-ablation on the rewarmed heart may be another way to assess adequate lesion creation but this too can be challenging to accomplish in the OR setting. Further research is required to help determine the optimal acute endpoints of SA. In this study, pre-discharge NIPS was performed in 10 of 20 patients and provided valuable prognostic information with regard to VT recurrence over the long-term: over 50% of patients who were non-inducible for VT on the NIPS experienced no further recurrent VT. Surgical endomyocardial resection of dense aneurysm has been shown to be effective in treating VT in patients with ICM due to prior myocardial infarction (where myocardial tissue within scar is non-viable and non-functioning) without causing significant detriment in cardiac function.2 However, it is important to recognize that in NICM patients where VT substrate is heterogeneous with poorly defined scar, if electrophysiologic guidance is inadequate, imprecise SA may result in destruction of potentially viable myocardium, possibly resulting in worsening of cardiac function. Aetiology of non-ischemic cardiomyopathy and outcome The heterogeneous mix of patients enabled us to explore any possible trends towards differences in outcome based on the underlying NICM aetiology. Although differences in VT-free survival did not reach statistical significance (likely due to small sample size), there was a non-significant trend towards improved survival and less VT recurrence in patients with ARVC or IDCM. This trend may have been on account of variations in scar distribution and/or the ability of cryoablation to effectively target the different substrates. The surprisingly high mortality rate after SA among patients with valvular heart disease may have been because all four of these patients underwent high-risk concomitant valve surgery (combined MV and TV repair in one, AV replacement alone in one, combined AV replacement with LVAD implant in one, and combined AV replacement, MV repair, and MAZE in one). The presence of severe valvular heart disease at baseline together with the additional surgical time likely contributed to the worse outcomes in this group. Outcomes in patients undergoing emergent surgical ablation It should be noted that in five patients, surgical intervention was performed emergently as a result of uncontrolled bleeding during challenging epicardial access. Three occurred in patients with NICM, one with HCM, and one with ARVC. Cause of uncontrolled bleeding was RV laceration in four and middle cardiac vein perforation in one. After the bleeding was controlled in the OR, direct visualization of the myocardium permitted SA. Site selection was based on information gathered in the electrophysiology lab, including VT morphology, and endocardial substrate characterization. Three of these patients remain alive and two have died in follow-up, one from pump failure and one from malignancy. Limitations There are several limitations to the present study. The sample size is small and heterogeneous. We acknowledge that since intra-operative voltage maps were not constructed and sustained VT was not induced and mapped during SA, the acute endpoints of successful ablation are not well-defined. Furthermore, in the few patients in whom adequate electrophysiologic data was unable to be gathered beforehand, the use of the 12-lead ECG morphology to localize VT exit sites along with gross visualization of scar may have been insufficient to accurately identify the critical isthmus. Also, half of the patients in our series did not undergo NIPS to assess for VT inducibility prior to hospital discharge after SA, which may have affected our study results. Furthermore, there may have been differences in the long-term cardiac management of our study population after SA could account for the differences in the observed outcomes. Finally, as has been previously discussed, in several patients in our series, SA was done in conjunction with other high-risk cardiac surgeries whereas in others this was performed emergently following a complication of percutaneous RFCA. Thus our findings, particularly the overall mortality rate, may not be representative of outcomes in clinically more stable, lower risk patients who are referred for elective SA. Conclusions In select patients with NICM experiencing VT refractory to conventional therapy or undergoing cardiac surgery for other indications, SA may be an option for potentially reducing the burden of ICD shocks during long-term follow-up. Detailed substrate and arrhythmia characterization in the electrophysiology lab prior to surgery, in conjunction with direct visualization of scar and prior radiofrequency ablation lesions in the operating room, is important for guiding SA in this unique patient population. Conflict of interest: none declared. References 1 Harken AH, Horowitz LN, Josephson ME. Surgical correction of recurrent sustained ventricular tachycardia following complete repair of tetralogy of Fallot. J Thorac Cardiovasc Surg  1980; 80: 779– 781. Google Scholar PubMed  2 Miller JM, Marchlinski FE, Harken AH, Hargrove WC, Josephson ME. Subendocardial resection for sustained ventricular tachycardia in the early period after acute myocardial infarction. Am J Cardiol  1985; 55: 980– 984. Google Scholar CrossRef Search ADS PubMed  3 Anter E, Hutchinson MD, Deo R, Haqqani HM, Callans DJ, Gerstenfeld EP et al.   Surgical ablation of refractory ventricular tachycardia in patients with nonischemic cardiomyopathy. Circ Arrhythm Electrophysiol  2011; 4: 494– 500. Google Scholar CrossRef Search ADS PubMed  4 Hsia HH, Callans DJ, Marchlinski FE. Characterization of endocardial electrophysiological substrate in patients with nonischemic cardiomyopathy and monomorphic ventricular tachycardia. Circulation  2003; 108: 704– 710. Google Scholar CrossRef Search ADS PubMed  5 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: 1288– 1296. Google Scholar CrossRef Search ADS PubMed  6 Sosa E, Scanavacca M, d'Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol  1996; 7: 531– 536. Google Scholar CrossRef Search ADS PubMed  7 Hutchinson MD, Gerstenfeld EP, Desjardins B, Bala R, Riley MP, Garcia FC et al.   Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Circ Arrhythm Electrophysiol  2011; 4: 49– 55. Google Scholar CrossRef Search ADS PubMed  8 Polin GM, Haqqani H, Tzou W, Hutchinson MD, Garcia FC, Callans DJ et al.   Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm  2011; 8: 76– 83. Google Scholar CrossRef Search ADS PubMed  9 Cano O, Hutchinson M, Lin D, Garcia F, Zado E, Bala R et al.   Electroanatomic substrate and ablation outcome for suspected epicardial ventricular tachycardia in left ventricular nonischemic cardiomyopathy. J Am Coll Cardiol  2009; 54: 799– 808. Google Scholar CrossRef Search ADS PubMed  10 Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR, Natterson PD et al.   Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation  1993; 88: 1647– 1670. Google Scholar CrossRef Search ADS PubMed  11 Jauregui-Abularach ME, Campos B, Betensky BP, Michele J, Gerstenfeld EP. Comparison of epicardial cryoablation and irrigated radiofrequency ablation in a Swine infarct model. J Cardiovasc Electrophysiol  2012; 23: 1016– 1023. Google Scholar CrossRef Search ADS PubMed  12 D'Avila A, Aryana A, Thiagalingam A, Holmvang G, Schmidt E, Gutierrez P et al.   Focal and linear endocardial and epicardial catheter-based cryoablation of normal and infarcted ventricular tissue. Pacing Clin Electrophysiol  2008; 31: 1322– 1331. Google Scholar CrossRef Search ADS PubMed  13 Hashimoto K, Watanabe I, Okumura Y, Ohkubo K, Ashino S, Kofune M et al.   Comparison of endocardial and epicardial lesion size following large-tip and extra-large-tip transcatheter cryoablation. Circ J  2009; 73: 1619– 1626. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Europace Oxford University Press

Loading next page...
 
/lp/ou_press/long-term-outcome-of-surgical-cryoablation-for-refractory-ventricular-q5JcxRjAo0
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com.
ISSN
1099-5129
eISSN
1532-2092
D.O.I.
10.1093/europace/eux029
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Limited data exist on the long-term outcome of patients (pts) with non-ischemic cardiomyopathy (NICM) and ventricular tachycardia (VT) refractory to conventional therapies undergoing surgical ablation (SA). We aimed to investigate the long-term survival and VT recurrence in NICM pts with VT refractory to radiofrequency catheter ablation (RFCA) who underwent SA. Methods and results Consecutive pts with NICM and VT refractory to RFCA who underwent SA were included. VT substrate was characterized in the electrophysiology lab and targeted by RFCA. During SA, previous RFCA lesions/scars were identified and targeted with cryoablation (CA; 3 min/lesion; target −150 °C). Follow-up comprised office visits, ICD interrogations and the social security death index. Twenty consecutive patients with NICM who underwent SA (age 53 ± 16 years, 18 males, LVEF 41 ± 20%; dilated CM = 9, arrhythmogenic right ventricular CM = 3, hypertrophic CM = 2, valvular CM = 4, and mixed CM = 2) were studied. Percutaneous mapping/ablation in the electrophysiology lab was performed in 18 and 2 pts had primary SA. During surgery, 4.9 ± 4.0 CA lesions/pt were delivered to the endocardium (2) and epicardium (11) or both (7). VT-free survival was 72.5% at 1 year and over 43 ± 31 months (mos) (range 1–83mos), there was only one arrhythmia-related death. There was a significant reduction in ICD shocks in the 3-mos preceding SA vs. the entire follow-up period (6.6 ± 4.9 vs. 2.3 ± 4.3 shocks/pt, P = 0.001). Conclusion In select pts with NICM and VT refractory to RFCA, SA guided by pre-operative electrophysiological mapping and ablation may be a therapeutic option. Ventricular tachycardia , Non-ischemic cardiomyopathy , Surgical cryoablation What’s new? Limited data exists on the long-term outcome of patients with non-ischemic cardiomyopathy (NICM) and VT refractory to catheter ablation who undergo surgical cryoablation (SA). We examined outcomes after SA in 20 consecutive patients with NICM and VT At 1 year post-SA, the VT-free survival rate was 72.5% and there was a significant reduction in ICD shocks (6.6 ± 4.9 in the 3 months pre-SA vs. 2.3 ± 4.3 shocks per patient over maximum 83 months post-SA, P = 0.001). There was only one arrhythmia-related death over 43 ± 31 months follow-up. In select pts with NICM and VT refractory to catheter ablation, SA guided by pre-operative electrophysiological mapping and ablation may be a therapeutic option. Introduction Prior to the catheter ablation era, surgical ablation (SA) was used to treat ventricular tachyarrhythmias, particularly in patients with healed myocardial infarction where this technique achieved over 80% success in the long-term management of recurrent of ventricular tachycardia (VT).1,2 With the advent of implantable cardioverter-defibrillators and percutaneous catheter ablation over the past two decades, SA is seldom utilized to treat ventricular arrhythmias. However, in some cases, where the ventricular arrhythmias are refractory or inaccessible to catheter ablation, SA may be considered. While there is long-standing experience with the SA technique in patients with ischemic cardiomyopathy (ICM), there is a paucity of data on the operative strategy for SA and long-term outcomes with this technique in patients with non-ischemic cardiomyopathy (NICM). We previously published our single centre experience on SA in eight consecutive patients with NICM where a priori electroanatomic mapping and electrophysiologic study in the electrophysiology lab was used to help guide the ablation approach in the operating room.3 This study demonstrated the feasibility and medium term efficacy of this strategy for successfully treating refractory VT in patients with NICM. In the present study, we aimed to describe our evolving experience with SA in a cohort of NICM patients with sustained VT refractory to catheter ablation, or in NICM patients who were undergoing cardiac surgery for other indications. We also sought to investigate the long-term outcomes of these patients in terms of both survival and VT recurrence. Methods Patient selection Patients >18 years with NICM and VT who underwent SA at the University of Pennsylvania from 2007 to 15 were included. Patients with pure ICM who underwent SA were excluded. All participants provided written informed consent for both the ablation procedure as well as inclusion in our VT ablation registry which was approved by the University of Pennsylvania Health System’s Institutional Review Board. Electronic medical records were retrospectively reviewed for baseline characteristics, clinical, and demographic data (Table 1). Table 1 Baseline patient characteristics N  20  Age (years)  53 ± 16  Males (#)  18  LVEF (%)  41 ± 20  RVEF (%)  37 ± 5  ICD  20  Disease type     Idiopathic dilated cardiomyopathy  9   Hypertrophic cardiomyopathy  2   Right ventricular cardiomyopathy  3   Valvular cardiomyopathy  4   Mixed ischemic/NICM  2  Cardiac MRI  13   Mid-myocardial LGE  3   Subepicardial LGE  3   Septal LGE  6  Prior ablation (# pts)     Total number procedures  1.7 ± 1.0   ENDO procedures  1.4 ± 0.8   EPI procedures  0.5 ± 0.7  Antiarrhythmic drugs (# pts)     Amiodarone  10   Sotalol  3   Quinidine  2   Mexilitine  2   Class IC  2  Comorbidities     Chronic kidney disease  9   Diabetes  4   Obstructive sleep apnea  5   Hyperlipidemia  2   Chronic obstructive pulmonary disease  1   Hypertension  7   Atrial fibrillation  6   Peripheral vascular disease  2   Sudden cardiac arrest  1   Thyroid disease  1   Heart block  2  Time RFCA to surgery (days prior)  46 ± 104 (median 6 days)  N  20  Age (years)  53 ± 16  Males (#)  18  LVEF (%)  41 ± 20  RVEF (%)  37 ± 5  ICD  20  Disease type     Idiopathic dilated cardiomyopathy  9   Hypertrophic cardiomyopathy  2   Right ventricular cardiomyopathy  3   Valvular cardiomyopathy  4   Mixed ischemic/NICM  2  Cardiac MRI  13   Mid-myocardial LGE  3   Subepicardial LGE  3   Septal LGE  6  Prior ablation (# pts)     Total number procedures  1.7 ± 1.0   ENDO procedures  1.4 ± 0.8   EPI procedures  0.5 ± 0.7  Antiarrhythmic drugs (# pts)     Amiodarone  10   Sotalol  3   Quinidine  2   Mexilitine  2   Class IC  2  Comorbidities     Chronic kidney disease  9   Diabetes  4   Obstructive sleep apnea  5   Hyperlipidemia  2   Chronic obstructive pulmonary disease  1   Hypertension  7   Atrial fibrillation  6   Peripheral vascular disease  2   Sudden cardiac arrest  1   Thyroid disease  1   Heart block  2  Time RFCA to surgery (days prior)  46 ± 104 (median 6 days)  MRI-LGE, magnetic resonance imaging-late gadolinium enhancement; LVEF, left ventricular ejection fraction; RVEF, right ventricular ejection fraction; ENDO, endocardial; EPI, epicardial; ICD, implantable cardioverter defibrillator; RFCA, radiofrequency catheter ablation. Catheter mapping and ablation Patients were brought to the electrophysiology lab in the post-absorptive state and percutaneous femoral access was obtained. A retrograde transaortic approach was used to access the left ventricle (LV). A detailed electroanatomic map of the chamber of interest was created during sinus rhythm or right ventricular (RV) pacing. All mapping was performed with a 3.5 mm, open-irrigation–tip catheter (Navistar Thermocool, Biosense Webster, Diamond Bar, CA). In each case, contact electroanatomic mapping (CARTO, Biosense Webster, Diamond Bar, CA) was performed using a fill and colour threshold  ≤15 mm, to ensure homogeneous sampling and representation of the entire endocardial surface area. Particular care was taken to sample more densely the scar and adjoining region.4,5 Epicardial access for mapping and ablation was typically performed in cases where the 12-lead electrocardiogram, pre-procedural cardiac imaging, and/or assessment of bipolar vs. unipolar voltage abnormalities on the electro-anatomic maps suggested the presence of epicardial substrate. Epicardial access was obtained with the technique originally described by Sosa et al.6 The reference values for identifying low-amplitude endocardial and epicardial bipolar and unipolar electrograms were defined as per previously established criteria.5,7–9 Standard techniques were used during the electrophysiology study, including programmed extrastimulation, entrainment mapping, pacemapping, and activation mapping.10 If the VT was not well tolerated or not reproducibly initiated, detailed characterization of the underlying substrate was performed and a substrate based ablation approach was undertaken. Radiofrequency ablation was performed, extending from the border zone to the dense scar, while transecting critical components of the VT circuit. When these regions were in close proximity to anatomic boundaries (mitral or aortic valve), the lesion sets were extended to incorporate these inexcitable areas. Typical settings during lesion creation were power range of 20–50 W and maximum temperature of 45°C, for a total duration of 60–180 s to achieve an impedance drop of 10–18 Ohms. Study protocol For patients included in the study, the three-dimensional electroanatomic maps were analysed retrospectively offline for anatomic scar distribution. The extent of abnormal endocardial and epicardial bipolar voltage signals was quantified by measuring contiguous areas of abnormal electrograms, using the ‘area calculation’ tool available in the 3D mapping system. Procedural reports were reviewed for intra-procedural data, including VT morphology, VT localization, and acute procedural endpoints. Surgical ablation protocol Patients were taken to the operating room in the post-absorptive state and placed under general anaesthesia. Access to the myocardium was mostly achieved through a midline sternotomy in order to obtain full visualization of the cardiac surface. In certain cases, per the surgeon’s discretion, a less invasive approach via partial lower sternotomy was used. Typically, when performing median sternotomy, myocardial protection was achieved with cold cardioplegia after placing patients on cardio-pulmonary bypass. The relevant surfaces of the heart were carefully inspected for visual and palpable scar as well as previously delivered radiofrequency lesions and key anatomic landmarks (coronary arteries, phrenic nerve, valves; Figure 1). Comparison of the observed scar was made to the pre-acquired electroanatomic map and presumed location of the VT exit based on the 12-lead electrocardiogram when available. In select cases, where an endocardial approach was deemed appropriate, endocardial exposure was accomplished through the aortic valve via an aortotomy above the sinus of Valsalva or through the LV apex [in patients undergoing left ventricular assist device (LVAD) placement]. No additional electrophysiologic mapping was performed during the surgical procedure. Cryothermy was applied to the targets (previous radiofrequency lesions, visible scars, presumptive VT exit sites based on 12-lead morphology) using the Surgifrost Surgical Cryoablation System (Medtronic CryoCath LP, QC, Canada). This is an Argon gas based system consisting of a flexible metal probe with an adjustable insulation sheath that can be moulded to conform to the cardiac contours. Typical cryo applications extended for 3 min (including the thawing phase), achieving a minimum temperature of −150 °C. Lesion delivery was assessed in real time through visual whitening and palpable induration of the myocardial surface. Following cryoablation, any additional surgical intervention (valve repair/replacement, coronary artery by-pass, etc.) was performed as clinically indicated and then the hearts were allowed to rewarm. Patients were acutely recovered in the cardiothoracic surgical intensive care unit and then transferred to the step-down unit. Non-invasive programmed stimulation (NIPS) was performed prior to discharge at the treating physician’s discretion. Anti-arrhythmic drugs were continued at discharge. Figure 1 View largeDownload slide Correlation of data from the electrophysiology lab with surgical cryoablation. Case example of a 51-year old male with a history of NICM (LVEF 35%) who presented with sustained VT and recurrent ICD therapies. Two endocardial and two epicardial radiofrequency ablation procedures were performed targeting a right bundle VT (A-B) with repeated late VT terminations during RF application. The epicardial lesion set was limited by phrenic nerve capture and coronary artery distribution (C) near the VT exit site (star). After the patient experienced recurrent VT with >10 ICD shocks, a surgical approach was undertaken. In the OR, prior epicardial RF lesions were identified and surgical cryoablation was performed from the epicardial LV surface (D). The patient has remained VT free off antiarrhythmic drug therapy over 71 months of follow-up. Figure 1 View largeDownload slide Correlation of data from the electrophysiology lab with surgical cryoablation. Case example of a 51-year old male with a history of NICM (LVEF 35%) who presented with sustained VT and recurrent ICD therapies. Two endocardial and two epicardial radiofrequency ablation procedures were performed targeting a right bundle VT (A-B) with repeated late VT terminations during RF application. The epicardial lesion set was limited by phrenic nerve capture and coronary artery distribution (C) near the VT exit site (star). After the patient experienced recurrent VT with >10 ICD shocks, a surgical approach was undertaken. In the OR, prior epicardial RF lesions were identified and surgical cryoablation was performed from the epicardial LV surface (D). The patient has remained VT free off antiarrhythmic drug therapy over 71 months of follow-up. Follow-up After discharge from the hospital, patients were subsequently followed through routine clinic visits at 6 weeks and then every 6 months. Survival and VT episodes were documented during office visits, remote and in-office ICD interrogations and through telephone calls. Mortality was also assessed through the Social Security Death Index. ICD programming and modification or discontinuation of the antiarrhythmic regimen was left to the discretion of the treating electrophysiologist. To account for multiple ICD therapies or electrical storm, all ICD interventions that occurred within a 24 h period were logged as a single event. The total number of ICD shocks was tabulated separately for comparison. There was no blanking period post-operatively. Statistical analysis Continuous variables are reported as mean ± standard deviations. Categorical data are presented as number of cases or percentages. For comparison of continuous variables between two groups, a paired Student’s t-test was performed for normally distributed variables. The Fisher’s exact test was used to analyse categorical data. A P < 0.05 was considered statistically significant. Kaplan–Meier survival curves were generated for all-cause death, death/transplant, VT recurrence post-SA, and VT recurrence after SA or repeat catheter ablation. Patient follow-up data were censored for either transplant or the time of last follow-up. Statistical analyses were performed on SPSSv21.0 software (SPSS, Chicago, IL). Results Over the study period (2007–15), 608 patients with NICM and/or mixed cardiomyopathy underwent catheter ablation of sustained VT at our institution. Eighteen of these patients required SA due to VT being refractory to ≥1 catheter ablation attempt(s) while two additional patients underwent primary SA (concomitant with aortic valve replacement in one patient and LVAD implantation in the other patient). Thus, the final study cohort comprised 20 patients (3.3% of patients with NICM undergoing VT ablation at our institution during the study period). In 15 patients (75%) SA was a planned procedure and in 5 patients (25%) it was performed on an emergent basis due to cardiac tamponade complicating radiofrequency catheter ablation (RFCA) procedures. The cause for cardiac tamponade was RV laceration during epicardial access in four patients and middle cardiac vein laceration in one. Eleven patients (55%) underwent concomitant planned cardiac surgery (five valve surgeries, two epicardial pacing lead placements, one MAZE surgery, two LVAD implantations, and one single coronary vessel bypass). Identification of substrate and ventricular arrhythmias Results of voltage mapping in the electrophysiology lab are listed in Tables 2 and 3. Consistent with our prior observations in patients with NICM, LV unipolar voltage abnormalities were more extensive than bipolar endocardial abnormalities and epicardial bipolar scar was more prominent than the endocardial bipolar scar. Additionally, the distribution of late gadolinium enhancement (LGE) representing scar as seen on cardiac MRI for each patient can be found in Table 3. A total of 62 VTs were induced in the electrophysiology lab. Forty-four VTs (71%) were localized or targeted from the endocardium and 18 were localized or targeted from the epicardium (29%). The primary reasons for failure of catheter ablation were presence of septal or mid-myocardial substrate in seven (35%), serious complication requiring emergent surgery in five (25%), proximity of target site to phrenic nerve and/or coronary arteries in four (20%), and inability to adequately access epicardial substrate due to the presence of dense pericardial adhesions in three (15%). Table 2 Electroanatomic voltage map characteristics and electrophysiologic study Endocardial bipolar low voltage     Perivalvular  11   Septal  6   Apical  1   None  2  Bipolar scar area (cm2)  28 ± 25  Unipolar voltage abnormalities     Perivalvular  13   Septal  11   Apical  5   None  1  Unipolar scar area (cm2)  88 ± 80  Epicardial bipolar low voltage     Perivalvular  4  Epicardial scar area (cm2)  37 ± 37  VT morphologies (total #/pt)  3.4 ± 2.3  VT morphologies (#/pt at last case)  2.7 ± 1.5  VTs haemodynamically tolerated (#/pt)  0.8 ± 1.5  Endocardial bipolar low voltage     Perivalvular  11   Septal  6   Apical  1   None  2  Bipolar scar area (cm2)  28 ± 25  Unipolar voltage abnormalities     Perivalvular  13   Septal  11   Apical  5   None  1  Unipolar scar area (cm2)  88 ± 80  Epicardial bipolar low voltage     Perivalvular  4  Epicardial scar area (cm2)  37 ± 37  VT morphologies (total #/pt)  3.4 ± 2.3  VT morphologies (#/pt at last case)  2.7 ± 1.5  VTs haemodynamically tolerated (#/pt)  0.8 ± 1.5  VT, ventricular tachycardia. Table 3 Clinical, electrophysiological, and imaging details for patients treated with surgical cryoablation for VT Pt #  Age  Aetiology of NICM  LVEF (%)  # ENDO/EPI RFA procedures  # VTs at last RFA  Clinical VTs at last RFA  Reason for RFCA failure  EP substrate (ENDO)  EP substrate (unipolar)  EP substrate (EPI)  MRI substrate (LGE)  1  52  Valvular  30  1/1  4  VT1: RBRS CL 450 ms  Phrenic nerve location  Perivalvular LV  Septum (base), anterior and anterolateral (base to mid), and lateral (basal)  Anterolateral (base to mid)  Anterior (base to mid) and septum  VT2: bidirectional  VT3: RBRI CL 420 ms  VT4: RBI CL 427 ms  2  72  NICM  50  2/0  1  VT1: RBRI, CL 295  Dense pericardial adhesions  None  Inferoseptal (base)  N/A  N/A  3  55  HOCM  65  2/1  5  VT1: RBRI CL 440  Septal substrate  Lateral  Lateral  Perivalvular (base)  Transmural septum (base to mid), subendocardial lateral (base), inferolateral  VT2: RBRI early transition  VT3: RBLI  VT4: RBI CL 500  VT5: RBI CL 280  4  65  Mixed  10  1/0  3  VT1: LBLI CL 370  Septal substrate  Septum (base)  Septum (base), inferolateral (base), lateral (base), periaortic  N/A  N/A  VT2: RBRS CL 290  VT3: LBLI CL 260  5  58  HOCM  70  1/1  2  VT1: RBS CL 294    LV apex  Septum (base), LV apex, lateral  N/A  Mid-myocardial lateral (base), anterolateral papillary muscle, inferior (mid), anteroseptum  VT2: LBLS CL 234  6  52  NICM  30  2/2  1  VT1: RBLS CL 400  Left circumflex coronary artery and phrenic nerve location  Periaortic  Septum (base)  Lateral  Mid-myocardial to epicardial inferolateral (base), inferior  7  43  Valvular  20  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  N/A  8  75  NICM  35  1/0  1  VT1: RBRI CL 300  Mid-myocardial substrate  Perimitral, periaortic  Diffuse LV (except apex)  N/A  N/A  9  48  NICM  55  3/1  3  VT1: RBRS CL 275  Mid-myocardial substrate  Septum (base), periaortic  Septum (base to mid), perimitral, anterolateral (patchy)  N/A  Mid-myocardial septum (base), anterior (base), inferior (base)  VT2: RBRI CL 325  VT3: RBRI CL 400  10  69  NICM  15  2/0  2  VT1: RBLI CL 450    Periaortic, inferoseptum, inferior (mid, patchy)  Septum (base to apex), inferior, perivalvular.  N/A  N/A  VT2: RBRS  11  51  Valvular  60  1/0  1  VT1: RBLS  Septal substrate  None  None  N/A  Subepicardial inferoseptum (mid)  12  49  ARVC  65  2/0  4  VT1: LBLS CL 243    Inferolateral RV (base to mid)  Inferolateral RV (base to mid)  N/A  Anteroseptal RV (base, patchy)  VT2: LBLS CL 250  VT3: LBLS CL 210  VT4: LBLS CL 194  13  70  Valvular  25  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  Subendocardial inferolateral (base to mid)  14  22  NICM  56  1/1  1  VT1: LBLS  Phrenic nerve location  LV septum (mid, patchy)  None  RV free wall, LV apex  Subepicardial anterior (mid to apex), lateral (apex), RV insertion  15  66  Mixed  20  1/0  3  VT1: RBRI CL 630  Dense pericardial adhesions  Inferoseptum (base to mid), periaortic  Diffuse LV (except lateral wall)  N/A  N/A  VT2: LBRS CL 450  VT3: RBRI CL 445  16  58  NICM  20  1/0  5  VT1: RBRS CL 309    Septum and Inferoseptum (base to mid)  Diffuse LV  N/A  N/A  VT2: RBRS CL 241  VT3: RBS  VT4: RBRS  VT5: RBRS  17  28  ARVC  65  3/2  4  VT1: LBLS (no trans)  LAD and PDA coronary arteries  RV inferior, inferolateral, anterior  Entire RV  N/A  Anterolateral RV, inferior RV  VT2: LBLS (V5 trans)  VT3: LBLS (V5 trans)  VT4: LBLS (V6 trans)  18  59  NICM  25  1/0  2  VT1: RB CL 300  Mid-myocardal substrate  Lateral (base, patchy)  Septum (base), inferior (base to mid), anterolateral perivalvular (base, patchy)  N/A  Subendocardial lateral (base to mid)  VT2: RB CL 288  19  18  ARVC  60  2/0  5  VT1: LBL CL 240  Dense pericardial adhesions  Anterior RVOT free wall and peritricuspid  Diffuse RV  N/A  None  VT2: LBL CL 300  VT3: LBI CL 280  VT4: LBS CL 290  VT5: LBS CL 280  20  58  NICM  40  1/0  1  VT1: RBI CL 300  Mid-myocardial substrate  Periaortic  Periaortic, AMC and anterolateral (base)  N/A  None  Pt #  Age  Aetiology of NICM  LVEF (%)  # ENDO/EPI RFA procedures  # VTs at last RFA  Clinical VTs at last RFA  Reason for RFCA failure  EP substrate (ENDO)  EP substrate (unipolar)  EP substrate (EPI)  MRI substrate (LGE)  1  52  Valvular  30  1/1  4  VT1: RBRS CL 450 ms  Phrenic nerve location  Perivalvular LV  Septum (base), anterior and anterolateral (base to mid), and lateral (basal)  Anterolateral (base to mid)  Anterior (base to mid) and septum  VT2: bidirectional  VT3: RBRI CL 420 ms  VT4: RBI CL 427 ms  2  72  NICM  50  2/0  1  VT1: RBRI, CL 295  Dense pericardial adhesions  None  Inferoseptal (base)  N/A  N/A  3  55  HOCM  65  2/1  5  VT1: RBRI CL 440  Septal substrate  Lateral  Lateral  Perivalvular (base)  Transmural septum (base to mid), subendocardial lateral (base), inferolateral  VT2: RBRI early transition  VT3: RBLI  VT4: RBI CL 500  VT5: RBI CL 280  4  65  Mixed  10  1/0  3  VT1: LBLI CL 370  Septal substrate  Septum (base)  Septum (base), inferolateral (base), lateral (base), periaortic  N/A  N/A  VT2: RBRS CL 290  VT3: LBLI CL 260  5  58  HOCM  70  1/1  2  VT1: RBS CL 294    LV apex  Septum (base), LV apex, lateral  N/A  Mid-myocardial lateral (base), anterolateral papillary muscle, inferior (mid), anteroseptum  VT2: LBLS CL 234  6  52  NICM  30  2/2  1  VT1: RBLS CL 400  Left circumflex coronary artery and phrenic nerve location  Periaortic  Septum (base)  Lateral  Mid-myocardial to epicardial inferolateral (base), inferior  7  43  Valvular  20  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  N/A  8  75  NICM  35  1/0  1  VT1: RBRI CL 300  Mid-myocardial substrate  Perimitral, periaortic  Diffuse LV (except apex)  N/A  N/A  9  48  NICM  55  3/1  3  VT1: RBRS CL 275  Mid-myocardial substrate  Septum (base), periaortic  Septum (base to mid), perimitral, anterolateral (patchy)  N/A  Mid-myocardial septum (base), anterior (base), inferior (base)  VT2: RBRI CL 325  VT3: RBRI CL 400  10  69  NICM  15  2/0  2  VT1: RBLI CL 450    Periaortic, inferoseptum, inferior (mid, patchy)  Septum (base to apex), inferior, perivalvular.  N/A  N/A  VT2: RBRS  11  51  Valvular  60  1/0  1  VT1: RBLS  Septal substrate  None  None  N/A  Subepicardial inferoseptum (mid)  12  49  ARVC  65  2/0  4  VT1: LBLS CL 243    Inferolateral RV (base to mid)  Inferolateral RV (base to mid)  N/A  Anteroseptal RV (base, patchy)  VT2: LBLS CL 250  VT3: LBLS CL 210  VT4: LBLS CL 194  13  70  Valvular  25  0/0  N/Aa  N/A  No prior RFA  N/A  N/A  N/A  Subendocardial inferolateral (base to mid)  14  22  NICM  56  1/1  1  VT1: LBLS  Phrenic nerve location  LV septum (mid, patchy)  None  RV free wall, LV apex  Subepicardial anterior (mid to apex), lateral (apex), RV insertion  15  66  Mixed  20  1/0  3  VT1: RBRI CL 630  Dense pericardial adhesions  Inferoseptum (base to mid), periaortic  Diffuse LV (except lateral wall)  N/A  N/A  VT2: LBRS CL 450  VT3: RBRI CL 445  16  58  NICM  20  1/0  5  VT1: RBRS CL 309    Septum and Inferoseptum (base to mid)  Diffuse LV  N/A  N/A  VT2: RBRS CL 241  VT3: RBS  VT4: RBRS  VT5: RBRS  17  28  ARVC  65  3/2  4  VT1: LBLS (no trans)  LAD and PDA coronary arteries  RV inferior, inferolateral, anterior  Entire RV  N/A  Anterolateral RV, inferior RV  VT2: LBLS (V5 trans)  VT3: LBLS (V5 trans)  VT4: LBLS (V6 trans)  18  59  NICM  25  1/0  2  VT1: RB CL 300  Mid-myocardal substrate  Lateral (base, patchy)  Septum (base), inferior (base to mid), anterolateral perivalvular (base, patchy)  N/A  Subendocardial lateral (base to mid)  VT2: RB CL 288  19  18  ARVC  60  2/0  5  VT1: LBL CL 240  Dense pericardial adhesions  Anterior RVOT free wall and peritricuspid  Diffuse RV  N/A  None  VT2: LBL CL 300  VT3: LBI CL 280  VT4: LBS CL 290  VT5: LBS CL 280  20  58  NICM  40  1/0  1  VT1: RBI CL 300  Mid-myocardial substrate  Periaortic  Periaortic, AMC and anterolateral (base)  N/A  None  a Patients 7 and 13 underwent primary SA at the time of LVAD implantation and aortic valve replacement, respectively. Surgical ablation Surgical ablation was performed a median of 7 days after the last catheter ablation (range 0–368 days). Details from the SA procedure for each patient can be found in Table 4. The primary reasons for undergoing surgery were: VT not amenable to or refractory to RFCA in 12 (60%), emergency repair of catheter ablation complication in 5 (25%), LVAD implantation in 2 (10%), and aortic valve replacement in 1 (5%). Surgical ablation was performed via midline sternotomy in 18 patients and via partial lower sternotomy in 2 patients. Seventeen SAs were performed on-pump and three were done off-pump. Cross clamp time was 66 ± 29 min with a bypass time of 120 ± 39 min. An average of 4.9 ± 4.0 (median 3) cryoablation applications were delivered per patient and the mean cryothermy time was 17 ± 14 min. In 2 patients only the endocardium was targeted and in 11 patients only the epicardium was targeted. Seven patients had cryo-lesions delivered to both the endocardium and epicardium. Endocardial access for cryoablation was obtained via ascending aortotomy (seven patients), atriotomy (two patients), or stab puncture through the RV outflow tract (one patient). The phrenic nerve was deflected and coronary arteries were dissected and displaced to allow for targeted cryoablation of the substrate in the four patients who had previously failed catheter ablation for that reason. In all seven patients who had previously failed catheter ablation due to the presence of septal or mid-myocardial substrate, the surgeons were able to target this location and achieve palpable transmurality of lesions with cryoablation. Table 4 Surgical details for patients treated with surgical cryoablation for VT Pt #  Interval between last RFCA and SA (Days)  Main reason for surgery  On/off pump  Pump time (min)  Cross-clamp time (min)  Surgical access  Additional cardiac surgery  ENDO/EPI cryoablation  Endo access  Cryoablation location  1  13  VT, severe MR  On  132  72  MS  MV repair, TV repair, LV epicardial lead  ENDO+EPI  Stab puncture through RVOT  Anterior RV septum (ENDO) and RV lateral to LAD (EPI)  2  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  Evacuation of pericardial hematoma, RV laceration repair  EPI  –  LV basal lateral and basal inferior  3  17  VT, AF  On  189  111  MS  MAZE  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  4  13  Severe MR (MV ring dehiscence) and VT  On  159  110  MS  Redo MV repair, removal of epicardial adhesions, LV epicardial lead  ENDO+EPI  Incision through LA suture line  Posterolateral septum (ENDO); Posterior, inferior, and posterolateral (EPI)  5  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  Inferolateral and inferior portions of LV apex  6  31  VT  On  77  42  MS  None  ENDO+EPI  Ascending aortotomy  LV posterior wall (ENDO+EPI)  7  N/Aa  LVAD implantation, VT  On  152  66  MS  AVR, LVAD  EPI only  –  Epicardium overlying RV/LV  8  7  VT  On  155  34  MS  None  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  9  5  VT  On  110  59  MS  LV epicardial lead  ENDO+EPI  Ascending aortotomy  LV septum and free wall (ENDO); anterior and posterior LV near septum (EPI)  10  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  RV laceration repair, LV epicardial lead  EPI only  –  LV inferior wall  11  31  Severe AR, VT  On  125  96  MS  AVR due to severe AI from LCC perforation  ENDO only  Ascending aortotomy  LV inferoseptum  12  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  RV near acute marginal branch  13  N/Aa  Severe AS, severe MR, VT  On  113  74  MS  AVR, MV repair, MAZE  ENDO only  Ascending aortotomy, left atriotomy at interatrial groove  LV inferior wall  14  289  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  Near RV acute marginal, RV and LV on either side of PDA, LV inferolateral and inferoapex  15  6  VT, CAD  On  81  46  MS  CABG, epicardial LV lead  EPI only  –  LV base, adjacent to CS lead  16  41  Continued VT and need for LVAD implantation  On  62  17  MS  LVAD  EPI only  –  LV inferolateral  17  5  VT  On  N/A  N/A  MS  None  EPI only  –  RV aspect of LAD and RVOT; along basal TV; parallel to RPDA  18  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  MCV tear repair  EPI only  –  LV inferolateral and posterolateral  19  4  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  RVOT to PV; RV anterior free wall to RV inferior wall  20  368  Severe MR and VT  On  87  68  MS  MV repair  ENDO+EPI  Ascending aortotomy  LVOT (ENDO) and anterior to aortic valve (EPI)  Pt #  Interval between last RFCA and SA (Days)  Main reason for surgery  On/off pump  Pump time (min)  Cross-clamp time (min)  Surgical access  Additional cardiac surgery  ENDO/EPI cryoablation  Endo access  Cryoablation location  1  13  VT, severe MR  On  132  72  MS  MV repair, TV repair, LV epicardial lead  ENDO+EPI  Stab puncture through RVOT  Anterior RV septum (ENDO) and RV lateral to LAD (EPI)  2  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  Evacuation of pericardial hematoma, RV laceration repair  EPI  –  LV basal lateral and basal inferior  3  17  VT, AF  On  189  111  MS  MAZE  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  4  13  Severe MR (MV ring dehiscence) and VT  On  159  110  MS  Redo MV repair, removal of epicardial adhesions, LV epicardial lead  ENDO+EPI  Incision through LA suture line  Posterolateral septum (ENDO); Posterior, inferior, and posterolateral (EPI)  5  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  Inferolateral and inferior portions of LV apex  6  31  VT  On  77  42  MS  None  ENDO+EPI  Ascending aortotomy  LV posterior wall (ENDO+EPI)  7  N/Aa  LVAD implantation, VT  On  152  66  MS  AVR, LVAD  EPI only  –  Epicardium overlying RV/LV  8  7  VT  On  155  34  MS  None  ENDO+EPI  Ascending aortotomy  LV base (ENDO+EPI)  9  5  VT  On  110  59  MS  LV epicardial lead  ENDO+EPI  Ascending aortotomy  LV septum and free wall (ENDO); anterior and posterior LV near septum (EPI)  10  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  RV laceration repair, LV epicardial lead  EPI only  –  LV inferior wall  11  31  Severe AR, VT  On  125  96  MS  AVR due to severe AI from LCC perforation  ENDO only  Ascending aortotomy  LV inferoseptum  12  0  Emergent; cardiac tamponade  Off  N/A  N/A  MS  RV laceration repair  EPI only  –  RV near acute marginal branch  13  N/Aa  Severe AS, severe MR, VT  On  113  74  MS  AVR, MV repair, MAZE  ENDO only  Ascending aortotomy, left atriotomy at interatrial groove  LV inferior wall  14  289  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  Near RV acute marginal, RV and LV on either side of PDA, LV inferolateral and inferoapex  15  6  VT, CAD  On  81  46  MS  CABG, epicardial LV lead  EPI only  –  LV base, adjacent to CS lead  16  41  Continued VT and need for LVAD implantation  On  62  17  MS  LVAD  EPI only  –  LV inferolateral  17  5  VT  On  N/A  N/A  MS  None  EPI only  –  RV aspect of LAD and RVOT; along basal TV; parallel to RPDA  18  0  Emergent; cardiac tamponade  On  N/A  N/A  MS  MCV tear repair  EPI only  –  LV inferolateral and posterolateral  19  4  VT  On  N/A  N/A  Partial lower sternotomy  None  EPI only  –  RVOT to PV; RV anterior free wall to RV inferior wall  20  368  Severe MR and VT  On  87  68  MS  MV repair  ENDO+EPI  Ascending aortotomy  LVOT (ENDO) and anterior to aortic valve (EPI)  Follow-up Early/in-hospital follow-up Eighteen of twenty patients (90%) survived to hospital discharge. Cause of death in the two patients who died before discharge was recurrent VT/VF shortly after SA in one patient and sepsis in the other. Mean hospital duration was 28 ± 34 days with a mean time from surgery to discharge of 22 ± 32 days (median 9.5 days). Patients spent a mean of 1.5 ± 1.1 days on mechanical ventilation after the surgery. Non-invasive programmed stimulation was performed in 10 (50%) patients during the index hospitalization. In one patient this was performed in the OR after the heart was rewarmed and in nine patients it was performed pre-discharge at an average of 9 ± 12 days after SA. In 9 of 10 patients (90%), VT was non-inducible at NIPS. Long-term follow-up Follow-up details for each patient are listed in Table 5. Long-term follow-up data revealed a 72.5% VT-free survival at 1-year, 59.3% VT-free survival at 2-years, and 34.6% VT-free survival at 5-years follow-up (Figure 2A). Over a mean follow-up duration of 43 ± 31 months (median 20 months), eight patients died (four pump failure, one refractory VT/VF, two malignancy, one septic shock). Of the 12 living patients, 1 underwent orthotopic heart transplant. Figure 2B shows survival free from death and Figure 2C shows freedom from death or transplant after SA. Of the 11 remaining patients, 5 experienced recurrent VT. Mean time to VT recurrence was 11 ± 15 months. In these five patients freedom from VT was achieved by an antiarrhythmic drug in three, left ganglionic sympathectomy in one, and repeat catheter ablation in four (median 13.8 months after SA). Figure 2D depicts freedom from VT after last surgical/catheter ablation procedure (allowing for repeat catheter ablation attempts after SA). Ten patients were on antiarrhythmic drugs at the last clinic follow-up visit (7 Amiodarone, 1 Amiodarone and Mexiletine, 1 Sotalol). Of the nine patients who were non-inducible for VT at NIPS, five (56%) experienced no subsequent VT recurrences. Table 5 Acute and long-term outcomes after surgical cryoablation for VT Pt #  Age  Year  Aetiology of NICM  VT storm pre-SA  ICD shocks per month in 3 months pre-SA  NIPS post-SA  Inducible at post-SA NIPS  VT storm post- SA  ICD shocks per month long-term post-SA  AAD drug post-SA  Duration of follow-up (months)  Outcome  1  52  2007  Valvular  Yes  2.3  Yes  No  No  0  Beta- blockerQuinidine  3  Deceased (pump failure)  2  72  2008  NICM  No  2.3  Yes  No  No  0  Beta-blocker  78  Alive  3  55  2008  HOCM  No  1  No  –  No  0  Sotalol  76  Alive  4  65  2008  Mixed  Yes  2.7  No  –  No  0  Amiodarone  18  Deceased (pump failure)  5  58  2008  HOCM  No  1  No  –  No  0.03  Amiodarone  72  Repeat catheter ablation  Deceased (malignancy)  6  52  2009  NICM  No  4  Yes  No  No  0  Beta-blocker  71  Alive  7  43  2009  Valvular  Yes  3.7  No  –  No  0.1  Amiodarone  35  Transplant  8  75  2009  NICM  Yes  5.3  No  –  Yes  3.7  Amiodarone  5  Repeat catheter ablation  Sympathectomy  Deceased (sepsis)  9  48  2009  NICM  No  2  Yes  Yes, non-clinical only  No  0.06  Amiodarone  64  Alive  10  69  2010  NICM  No  1.7  No  –  No  0  Amiodarone  2  Deceased (pump failure)  11  51  2010  Valvular  No  –  Yes  No  No  0  Beta-blocker  48  Alive  12  49  2011  ARVC  Yes  0.3  Yes  No  No  0  Beta-blocker  40  Alive  13  70  2011  Valvular  No  0.3  No  –  No  0.2  Beta-blocker  23  Repeat catheter ablation  Deceased  (Pump failure)  14  22  2012  NICM  No  0  Yes  No  No  0.03  Beta-blocker  33  Repeat catheter ablation  Alive  15  66  2012  Mixed  Yes  4.3  No  –  No  2  Beta-blocker  0  Deceased (VT/VF)  16  58  2013  NICM  No  2.3  No  –  No  0  Amiodarone  7  Deceased (malignancy)  17  28  2013  ARVC  No  0.7  Yes  No  No  0  Beta-blocker  14  Alive  18  59  2014  NICM  No  4.7  No  –  No  0.3  Amiodarone  8  Alive  Mexilitine  19  18  2014  ARVC  No  2.7  Yes  No  No  2.8  Beta-blocker  1  Sympathectomy  Alive  20  58  2015  NICM  No  0.3  No  –  No  0  Beta-blocker  1  Alive  Pt #  Age  Year  Aetiology of NICM  VT storm pre-SA  ICD shocks per month in 3 months pre-SA  NIPS post-SA  Inducible at post-SA NIPS  VT storm post- SA  ICD shocks per month long-term post-SA  AAD drug post-SA  Duration of follow-up (months)  Outcome  1  52  2007  Valvular  Yes  2.3  Yes  No  No  0  Beta- blockerQuinidine  3  Deceased (pump failure)  2  72  2008  NICM  No  2.3  Yes  No  No  0  Beta-blocker  78  Alive  3  55  2008  HOCM  No  1  No  –  No  0  Sotalol  76  Alive  4  65  2008  Mixed  Yes  2.7  No  –  No  0  Amiodarone  18  Deceased (pump failure)  5  58  2008  HOCM  No  1  No  –  No  0.03  Amiodarone  72  Repeat catheter ablation  Deceased (malignancy)  6  52  2009  NICM  No  4  Yes  No  No  0  Beta-blocker  71  Alive  7  43  2009  Valvular  Yes  3.7  No  –  No  0.1  Amiodarone  35  Transplant  8  75  2009  NICM  Yes  5.3  No  –  Yes  3.7  Amiodarone  5  Repeat catheter ablation  Sympathectomy  Deceased (sepsis)  9  48  2009  NICM  No  2  Yes  Yes, non-clinical only  No  0.06  Amiodarone  64  Alive  10  69  2010  NICM  No  1.7  No  –  No  0  Amiodarone  2  Deceased (pump failure)  11  51  2010  Valvular  No  –  Yes  No  No  0  Beta-blocker  48  Alive  12  49  2011  ARVC  Yes  0.3  Yes  No  No  0  Beta-blocker  40  Alive  13  70  2011  Valvular  No  0.3  No  –  No  0.2  Beta-blocker  23  Repeat catheter ablation  Deceased  (Pump failure)  14  22  2012  NICM  No  0  Yes  No  No  0.03  Beta-blocker  33  Repeat catheter ablation  Alive  15  66  2012  Mixed  Yes  4.3  No  –  No  2  Beta-blocker  0  Deceased (VT/VF)  16  58  2013  NICM  No  2.3  No  –  No  0  Amiodarone  7  Deceased (malignancy)  17  28  2013  ARVC  No  0.7  Yes  No  No  0  Beta-blocker  14  Alive  18  59  2014  NICM  No  4.7  No  –  No  0.3  Amiodarone  8  Alive  Mexilitine  19  18  2014  ARVC  No  2.7  Yes  No  No  2.8  Beta-blocker  1  Sympathectomy  Alive  20  58  2015  NICM  No  0.3  No  –  No  0  Beta-blocker  1  Alive  Figure 2 View largeDownload slide Kaplan–Meier curves for patients with a NICM who underwent surgical cryoablation for refractory VT at our institution. (A) Survival free of VT after SA. (B) Survival free of death from any cause. (C) Survival free from death or heart transplantation. (D) Survival free of VT (after SA or last catheter ablation). Figure 2 View largeDownload slide Kaplan–Meier curves for patients with a NICM who underwent surgical cryoablation for refractory VT at our institution. (A) Survival free of VT after SA. (B) Survival free of death from any cause. (C) Survival free from death or heart transplantation. (D) Survival free of VT (after SA or last catheter ablation). In terms of total ICD interventions, there was a non-significant trend towards a reduction of total VT events from the 3 months pre-SA compared to the entire post-SA follow-up period (maximum 83 months; 5.7 ± 6.0 vs. 3.2 ± 5.9 events/pt, median 3.0 vs. 1.0 events, P = 0.11; Figure 3). There was a significant reduction in ICD shocks from the 3 months pre-SA to the first 3 months post-SA (mean 6.6 ± 4.9 vs. 1.7 ± 4.3 shocks/pt, respectively; median 7.0 vs. 0.0 events, respectively; P < 0.001) as well as the entire post-SA follow-up duration of 83 months (6.6 ± 4.9 vs. 2.3 ± 4.3 shocks/pt, respectively; median 7.0 vs. 0.0 shocks/pt, respectively; P = 0.001; Figure 3). Of the six patients in our cohort who had experienced VT storm in the interval between their last RFCA attempt and the SA, only one had recurrent VT storm after SA. Figure 3 View largeDownload slide VT events and ICD shocks in patients with NICM undergoing surgical cryoablation for refractory VT. There was a significant reduction in ICD shocks per patient and a non-significant trend towards a reduction in total VT events. Figure 3 View largeDownload slide VT events and ICD shocks in patients with NICM undergoing surgical cryoablation for refractory VT. There was a significant reduction in ICD shocks per patient and a non-significant trend towards a reduction in total VT events. Overall survival based on the aetiology of underlying cardiomyopathy were as follows: arrhythmogenic right ventricular cardiomyopathy (ARVC)—100%, idiopathic dilated cardiomyopathy (IDCM)—67%, hypertrophic cardiomyopathy (HCM)—50%, and valvular cardiomyopathy—25%. These differences, however, were not significant (P = 0.13). There were no significant differences in VT recurrence rates between the subtypes of NICM (ARVC: 33%, IDCM: 44%, valvular cardiomyopathy: 75%, and HCM: 100%; P = 0.32). Discussion While RFCA is an effective treatment option for many patients with VT and NICM, standard techniques for endocardial and epicardial RFCA may be inadequate in a minority of patients. Common causes of failure of percutaneous RFCA for VT include prohibitive proximity of target site to coronary arteries or phrenic nerve, failure to gain percutaneous epicardial access due to adhesions from prior cardiac surgery, and the presence of mid-myocardial or septal substrate. We report our institution’s long-term experience of SA in a consecutive group of NICM patients who suffered from recurrent, symptomatic VT refractory to medical therapy, and percutaneous RFCA. For some of these patients who were not considered transplant candidates, this was a treatment option of last resort. In other patients this was offered during open heart surgery for other indications. In 90% of cases, SA was guided by information gathered in the electrophysiology lab utilizing electroanatomic, activation and pace mapping to characterize the underlying substrate and define critical components of the re-entrant VT circuit. Our long-term outcomes in this critically ill patient population showed a 72.5% freedom from VT at 1 year and a significant reduction in ICD shocks over the long-term. The use of anti-arrhythmic drugs was also reduced from 95 to 50% over the course of the extended follow-up. Overall survival was 60% over 7 years and arrhythmic death was rare. Ablation technique and endpoints Surgical cryoablation for VT in patients with NICM is rarely needed and was used in <5% of our total ablation population. In these unusual cases, we speculate several reasons why SA may have been effective. First, our strategy of utilizing data from the electrophysiology lab for guidance together with surgical access and cold cardioplegia enabled us to more precisely target the VT substrate in the OR under direct visualization of all cardiac surfaces. Secondly, scar is frequently mid-myocardial in patients with NICM. The current RFCA technology may not be able to create deep enough lesions in some patients. This is particularly true when targeting the basal epicardial region where the myocardium is thick and the presence of fat may hinder adequate lesion creation from the epicardium. Delivering cryoablation in the OR under cold cardioplegia may allow for creation of larger lesions.11–13 Technical issues such as catheter stability and contact force during percutaneous catheter ablation are not encountered during surgical cryoablation. Furthermore, structures such as epicardial coronary arteries and the phrenic nerve can be clearly visualized in the OR and manually deflected if ablation is required in their vicinity. Also, dense pericardial adhesions can make percutaneous epicardial access challenging. Ideally, we would have preferred to induce and map the VT in the OR similar to the approach that is used in the electrophysiology lab. However, this is not always feasible for a variety of reasons. First, it requires similar mapping environment and tools as in the electrophysiology lab. Secondly, it can be difficult to induce and sustain VT under cold cardioplegia. The latter scenario also poses a challenge to assessing the intraprocedural success of the ablation procedure. In our series, the surgeons used alternate intra-operative endpoints to assess for adequacy of lesion creation in the OR including visualizing tissue whitening at the surface where the lesion is delivered and feeling for palpable induration on the surface opposing the site of lesion creation (epicardial surface for endocardial lesions and vice-versa). Another criteria that we intend to test in future cases to validate effective lesion delivery in the OR is local inexcitability with high output pacing (10–50 mA × 2 ms). Additional assessment of procedural efficacy may be to perform programmed stimulation after warming the heart. This can be challenging especially if additional cardiac surgery is being contemplated immediately following SA as was the case in more than half of our study population. In the future, defining myocardial voltage changes post-ablation on the rewarmed heart may be another way to assess adequate lesion creation but this too can be challenging to accomplish in the OR setting. Further research is required to help determine the optimal acute endpoints of SA. In this study, pre-discharge NIPS was performed in 10 of 20 patients and provided valuable prognostic information with regard to VT recurrence over the long-term: over 50% of patients who were non-inducible for VT on the NIPS experienced no further recurrent VT. Surgical endomyocardial resection of dense aneurysm has been shown to be effective in treating VT in patients with ICM due to prior myocardial infarction (where myocardial tissue within scar is non-viable and non-functioning) without causing significant detriment in cardiac function.2 However, it is important to recognize that in NICM patients where VT substrate is heterogeneous with poorly defined scar, if electrophysiologic guidance is inadequate, imprecise SA may result in destruction of potentially viable myocardium, possibly resulting in worsening of cardiac function. Aetiology of non-ischemic cardiomyopathy and outcome The heterogeneous mix of patients enabled us to explore any possible trends towards differences in outcome based on the underlying NICM aetiology. Although differences in VT-free survival did not reach statistical significance (likely due to small sample size), there was a non-significant trend towards improved survival and less VT recurrence in patients with ARVC or IDCM. This trend may have been on account of variations in scar distribution and/or the ability of cryoablation to effectively target the different substrates. The surprisingly high mortality rate after SA among patients with valvular heart disease may have been because all four of these patients underwent high-risk concomitant valve surgery (combined MV and TV repair in one, AV replacement alone in one, combined AV replacement with LVAD implant in one, and combined AV replacement, MV repair, and MAZE in one). The presence of severe valvular heart disease at baseline together with the additional surgical time likely contributed to the worse outcomes in this group. Outcomes in patients undergoing emergent surgical ablation It should be noted that in five patients, surgical intervention was performed emergently as a result of uncontrolled bleeding during challenging epicardial access. Three occurred in patients with NICM, one with HCM, and one with ARVC. Cause of uncontrolled bleeding was RV laceration in four and middle cardiac vein perforation in one. After the bleeding was controlled in the OR, direct visualization of the myocardium permitted SA. Site selection was based on information gathered in the electrophysiology lab, including VT morphology, and endocardial substrate characterization. Three of these patients remain alive and two have died in follow-up, one from pump failure and one from malignancy. Limitations There are several limitations to the present study. The sample size is small and heterogeneous. We acknowledge that since intra-operative voltage maps were not constructed and sustained VT was not induced and mapped during SA, the acute endpoints of successful ablation are not well-defined. Furthermore, in the few patients in whom adequate electrophysiologic data was unable to be gathered beforehand, the use of the 12-lead ECG morphology to localize VT exit sites along with gross visualization of scar may have been insufficient to accurately identify the critical isthmus. Also, half of the patients in our series did not undergo NIPS to assess for VT inducibility prior to hospital discharge after SA, which may have affected our study results. Furthermore, there may have been differences in the long-term cardiac management of our study population after SA could account for the differences in the observed outcomes. Finally, as has been previously discussed, in several patients in our series, SA was done in conjunction with other high-risk cardiac surgeries whereas in others this was performed emergently following a complication of percutaneous RFCA. Thus our findings, particularly the overall mortality rate, may not be representative of outcomes in clinically more stable, lower risk patients who are referred for elective SA. Conclusions In select patients with NICM experiencing VT refractory to conventional therapy or undergoing cardiac surgery for other indications, SA may be an option for potentially reducing the burden of ICD shocks during long-term follow-up. Detailed substrate and arrhythmia characterization in the electrophysiology lab prior to surgery, in conjunction with direct visualization of scar and prior radiofrequency ablation lesions in the operating room, is important for guiding SA in this unique patient population. Conflict of interest: none declared. References 1 Harken AH, Horowitz LN, Josephson ME. Surgical correction of recurrent sustained ventricular tachycardia following complete repair of tetralogy of Fallot. J Thorac Cardiovasc Surg  1980; 80: 779– 781. Google Scholar PubMed  2 Miller JM, Marchlinski FE, Harken AH, Hargrove WC, Josephson ME. Subendocardial resection for sustained ventricular tachycardia in the early period after acute myocardial infarction. Am J Cardiol  1985; 55: 980– 984. Google Scholar CrossRef Search ADS PubMed  3 Anter E, Hutchinson MD, Deo R, Haqqani HM, Callans DJ, Gerstenfeld EP et al.   Surgical ablation of refractory ventricular tachycardia in patients with nonischemic cardiomyopathy. Circ Arrhythm Electrophysiol  2011; 4: 494– 500. Google Scholar CrossRef Search ADS PubMed  4 Hsia HH, Callans DJ, Marchlinski FE. Characterization of endocardial electrophysiological substrate in patients with nonischemic cardiomyopathy and monomorphic ventricular tachycardia. Circulation  2003; 108: 704– 710. Google Scholar CrossRef Search ADS PubMed  5 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: 1288– 1296. Google Scholar CrossRef Search ADS PubMed  6 Sosa E, Scanavacca M, d'Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol  1996; 7: 531– 536. Google Scholar CrossRef Search ADS PubMed  7 Hutchinson MD, Gerstenfeld EP, Desjardins B, Bala R, Riley MP, Garcia FC et al.   Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Circ Arrhythm Electrophysiol  2011; 4: 49– 55. Google Scholar CrossRef Search ADS PubMed  8 Polin GM, Haqqani H, Tzou W, Hutchinson MD, Garcia FC, Callans DJ et al.   Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm  2011; 8: 76– 83. Google Scholar CrossRef Search ADS PubMed  9 Cano O, Hutchinson M, Lin D, Garcia F, Zado E, Bala R et al.   Electroanatomic substrate and ablation outcome for suspected epicardial ventricular tachycardia in left ventricular nonischemic cardiomyopathy. J Am Coll Cardiol  2009; 54: 799– 808. Google Scholar CrossRef Search ADS PubMed  10 Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR, Natterson PD et al.   Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation  1993; 88: 1647– 1670. Google Scholar CrossRef Search ADS PubMed  11 Jauregui-Abularach ME, Campos B, Betensky BP, Michele J, Gerstenfeld EP. Comparison of epicardial cryoablation and irrigated radiofrequency ablation in a Swine infarct model. J Cardiovasc Electrophysiol  2012; 23: 1016– 1023. Google Scholar CrossRef Search ADS PubMed  12 D'Avila A, Aryana A, Thiagalingam A, Holmvang G, Schmidt E, Gutierrez P et al.   Focal and linear endocardial and epicardial catheter-based cryoablation of normal and infarcted ventricular tissue. Pacing Clin Electrophysiol  2008; 31: 1322– 1331. Google Scholar CrossRef Search ADS PubMed  13 Hashimoto K, Watanabe I, Okumura Y, Ohkubo K, Ashino S, Kofune M et al.   Comparison of endocardial and epicardial lesion size following large-tip and extra-large-tip transcatheter cryoablation. Circ J  2009; 73: 1619– 1626. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com.

Journal

EuropaceOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off