Author + information
- Received February 28, 2017
- Revision received May 30, 2017
- Accepted May 31, 2017
- Published online December 4, 2017.
- Joshua D. Moss, MDa,∗ (, )
- Erin E. Flatley, RN, MSN, ANP-BCb,
- Andrew D. Beaser, MDb,
- John H. Shin, MDc,
- Hemal M. Nayak, MDb,
- Gaurav A. Upadhyay, MDb,
- Martin C. Burke, DOd,
- Valluvan Jeevanandam, MDb,
- Nir Uriel, MDb and
- Roderick Tung, MDb
- aSection of Cardiac Electrophysiology, Division of Cardiology, University of California–San Francisco, San Francisco, California
- bHeart and Vascular Center, University of Chicago Medicine, Chicago, Illinois
- cMid-Atlantic Permanente Medical Group, Rockville, Maryland
- dCorVita Science Foundation, Chicago, Illinois
- ↵∗Address for correspondence:
Dr. Joshua D. Moss, Section of Cardiac Electrophysiology, University of California–San Francisco, 500 Parnassus Avenue, Box 1354, San Francisco, California 94143-1354.
Objectives This study sought to report mechanisms of ventricular tachycardia (VT) and outcomes of VT ablation in patients with a left ventricular assist device (LVAD) as destination therapy.
Background Continuous flow LVAD implantation plays a growing role in the management of end-stage heart failure, and VT is common. There are limited reports of VT ablation in patients with a destination LVAD.
Methods Patients with a continuous-flow LVAD referred for VT ablation from 2010 to 2016 were analyzed retrospectively. Baseline patient characteristics, procedural data, and clinical follow-up were evaluated. Arrhythmia-free survival was assessed.
Results Twenty-one patients (90% male, 62 ± 10 years) underwent catheter ablation of VT at a median of 191 days (interquartile range: 55 to 403 days) after LVAD implantation (15 HeartMate II, 6 HeartWare HVAD). Five patients (24%) had termination (n = 4) or slowing (n = 1) of VT with ablation near the apical inflow cannula, and 3 (14%) had bundle-branch re-entry. Freedom from recurrent VT among surviving patients was 64% at 1 year, with overall survival 67% at 1 year for patients without arrhythmia recurrence and 29% for patients with recurrence (p = 0.049). One patient had suspected pump thrombosis within 30 days of the ablation procedure, with no other major acute complications.
Conclusions In this relatively large, single-center experience of VT ablation in destination LVAD, freedom from recurrent VT and implantable cardioverter-defibrillator shocks was associated with improved 1-year survival. Bundle branch re-entry was more prevalent than anticipated, and cannula-adjacent VT was less common. This challenging population remains at risk for late pump thrombosis and mortality.
Left ventricular assist device (LVAD) therapy is increasingly used for treatment of advanced heart failure, having been shown to improve survival after eventual cardiac transplantation (1,2) as well as survival and quality of life as destination therapy in patients ineligible for transplantation (3,4). Earlier pulsatile-flow devices have largely given way to smaller and more durable continuous-flow devices, with improved outcomes (4–6).
Ventricular arrhythmias (VA) are common in patients with continuous-flow LVAD—particularly in those patients with a history of VA prior to LVAD implantation—and are associated with increased morbidity (7) and mortality (8,9). The benefits of catheter ablation for VA have been demonstrated across a wide spectrum of patients (10,11). Favorable outcomes of VA ablation in patients with continuous-flow LVAD have been published in several cohorts, including a single-center experience across the temporal spectrum of device technologies (12), a multicenter experience with the HeartMate II device (Thoratec, Pleasanton, California) (13), and several smaller cohorts (14–16).
We sought to describe the mechanisms, timing, and ablation outcomes in a relatively large single-center VA ablation experience in patients with a continuous-flow LVAD prescribed as destination therapy.
All patients with a continuous-flow LVAD who underwent an electrophysiology study and ablation procedure for VA at the University of Chicago Medicine were included. All relevant clinical and procedural data were collected retrospectively from the medical chart, procedure report, electrophysiology recording system, 3-dimensional mapping system, and implantable cardioverter-defibrillator (ICD) interrogations. Early VA was defined as sustained monomorphic ventricular tachycardia (VT) or ventricular fibrillation that occurred either within 30 days of LVAD implantation or during the same hospitalization as LVAD implantation. Retrospective analysis of the data was approved by the institutional review board.
The decision to proceed with VT ablation was made on a case-by-case basis at the discretion of the Heart Failure/Mechanical Circulatory Support team and the consulting electrophysiologist. None of the patients in this cohort were deemed eligible for cardiac transplantation. Ablation was most often considered for recurrent sustained VA requiring ICD therapy more than 1 month after LVAD implantation despite a trial of antiarrhythmic drug therapy. A smaller number of patients were also referred for VT storm, or concern for exacerbation of right ventricular (RV) failure in the setting of recurrent VT, occurring within 2 weeks after LVAD implantation. Suction events, with irritation of ventricular myocardium pulled into contact with the inflow cannula, could often be identified by stereotypical nonsustained and polymorphic bursts of VA, frequently correlating with transient changes in logged pump parameters. Whenever suspected, these were first treated with adjustment of diuretic dosing and/or LVAD pump speed, either empirically or guided by hemodynamic ramp testing (17), prior to consideration of ablative therapy.
Electrophysiology study and ablation
All procedures were performed at the University of Chicago Medical Center. Procedures were performed under conscious sedation or general endotracheal anesthesia per operator discretion. Anticoagulation with warfarin with target international normalized ratio (INR) 2 to 3 was continued uninterrupted or periprocedural anticoagulation was maintained with therapeutic heparin as needed. Heparin was also administered during all procedures to achieve and maintain a goal activated clotting time >300 s prior to any left-sided instrumentation. Catheters were advanced via femoral venous and/or arterial access, and LVAD controller parameters were continuously monitored. Temporary adjustments of LVAD pump speed were made as needed to facilitate retrograde aortic and/or transseptal access. Percutaneous epicardial access was not attempted in any of the cases.
Three-dimensional electroanatomic endocardial voltage maps were created using commercially available mapping systems (CARTO 3, Biosense Webster, Diamond Bar, California [n = 16]; Velocity, St. Jude Medical, Minneapolis, Minnesota [n = 4]; or Rhythmia, Boston Scientific, Natick, Massachusetts [n = 1]) and used to delineate areas of scar with standard low-voltage settings (<1.5 mV) (18) and abnormal electrical activation. Intracardiac echocardiography was to guide transseptal access when chosen, monitor catheter position and lesion formation, and visualize the LVAD inflow cannula. Cannula-adjacent VT was defined when targeted within 2 cm of the inflow cannula, and non-cannula-associated VT was defined when targeted within or near regions of scar that were remote and distinct from the inflow cannula.
Whenever clinically feasible, VA were induced with programmed stimulation to facilitate activation and/or entrainment mapping as appropriate. An isthmus was defined as a site that demonstrated concealed fusion with a post-pacing interval within 30 ms of the VT cycle length and a stimulus-to-QRS interval equal to electrogram-QRS (19). Pace-mapping was also used in the majority of procedures to help localize ablation in regions where matches with the targeted VT were seen, particularly at sites with longer stimulus-QRS latency (20). Regions of late activation or local conduction delay as evidenced by split, fractionated, or isolated late potentials were also tagged and targeted for ablation when feasible (21,22). Ablation was performed in all cases with irrigated radiofrequency (RF) energy (power 30 to 50 W, temperature limit of 42°C). The primary endpoint for ablation was termination and noninducibility of the clinical VT whenever possible; additional substrate modification and targeting of nonclinical VA were performed at the discretion of the primary operator.
Categorical variables are presented as ratios with corresponding percentages and were compared using the Fisher exact test given the small group size. Corresponding effect sizes are presented using the Cramer V statistic. Continuous variables are presented as mean ± SD or median (interquartile ranges [IQR]) when not normally distributed and compared using the Student t test or Mann-Whitney U test as appropriate. Corresponding effect sizes are presented using the Hedges g statistic or pseudo-r calculation, respectively. The Kaplan-Meier estimator was used for survival analysis. Statistical analyses were performed using SPSS Statistics (IBM, Chicago, Illinois). All analyses were 2-tailed, with statistical significance defined as p values of ≤0.05. Effect sizes were deemed negligible if <0.10, small if 0.10 to 0.30, moderate if 0.30 to 0.50, and large if >0.50.
From July 2010 to April 2016, 23 electrophysiological studies and ablation procedures were performed in 21 patients with a continuous-flow LVAD (HeartMate II = 15, HeartWare HVAD [Framingham, Massachusetts] = 6). Key clinical and procedural information about each patient as well as antiarrhythmic drug use and outcome by 1-year post-ablation are shown in Table 1. Overall, mean age at the time of ablation was 62.3 ± 10.3 years, and mean LV ejection fraction (EF) was 18.3 ± 5.2%. Over time, ablative therapy tended to be offered earlier in the treatment course as procedural experience with these complex patients increased and positive outcomes were seen. Patients later in the cohort were both less likely to be on multiple antiarrhythmic drugs prior to ablation and less likely to continue on high-dose amiodarone after ablation. However, no clear trend emerged with regard to timing of the actual ablation procedure relative to first occurrence of post-LVAD VA, likely due to the influence of other variables such as comorbidities, response to medications, psychological effects of ICD therapies, and care team attitudes.
Clinical characteristics of the cohort as categorized by diagnosis of any recurrent VA after ablation are presented in Table 2. No significant differences in age, EF, etiology of cardiomyopathy, type of LVAD, history of pre-LVAD ventricular arrhythmias, clinical presentation, markers of RV dysfunction, or medication use were found between those who had recurrence and those without recurrent VT after ablation.
Patients who had recurrent VA after ablation were twice as likely to have had no known VA prior to LVAD implantation (86% vs. 43%), and they were more likely to have VA present early (within 30 days) after LVAD implantation (86% vs. 57%). Median time from LVAD to first VA was 8 days (IQR: 2 to 29.5 days) in the group who ultimately had post-ablation recurrence, and 15.5 days (IQR: 4.75 to 222.75 days) in the group who had no recurrence after ablation. Median time from first VA presentation to an ablation procedure was similar in the 2 groups (62.5 vs. 74 days).
Procedural details and findings are shown for each patient in Table 1, and for the entire cohort as categorized by post-ablation recurrence in Table 3. A transseptal approach was employed in 76% of cases. Conscious sedation was used in 76% of patients. Endocardial scar was widely distributed throughout the LV (anterior in 75% of patients, inferior in 75%, septal in 81%, lateral in 44%, apical in 75%, and basal in 63%). A median of 3 different VT morphologies was induced per patient (IQR: 2 to 4). Activation and entrainment mapping were achieved in 90% of cases, despite many VT with rapid cycle lengths that might otherwise be poorly tolerated without the benefit of LVAD support; for the entire cohort, median VT cycle length was 340 ms (IQR: 280 to 375 ms).
The clinical VT was identified, induced, and mapped in 85% of patients and was more than twice as likely to have a right-bundle branch block morphology as a left-bundle one. Three patients had proven bundle-branch re-entrant ventricular tachycardia (BBRT), representing 14% of the overall cohort and 60% of those patients with a left-bundle branch block morphology. Typical left-bundle type was induced in all 3, with 1 patient displaying reverse BBRT as well (Figure 1). After successful ablation of the right bundle, no patients with BBRT had recurrent BBRT or ICD shocks through at least 1 year of post-ablation follow-up.
In patients who did not have recurrence after ablation, the median cycle length of all induced VT was significantly slower (350 vs. 310 ms, p = 0.03), although the cycle length of the clinical VT was similar. The clinical VT was identified more often and entrainment mapping used more often in those patients who did not have post-ablation recurrence. Termination of a clinical VT with RF achieved in 62% of patients. However, only 29% of patients were rendered completely noninducible after ablation, including only 1 of 3 with clinical BBRT.
Ablation of cannula-adjacent VT was performed in 35% of patients, 57% of whom had endocardial scar on electroanatomic mapping in the apical third of the LV (Figure 2). Ablation within or near endocardial scar remote from the inflow cannula was performed in 51% of patients (Figure 3). Termination or significant slowing of VT was noted with delivery of RF near the inflow cannula in 5 patients (24%), and twice as often in the group that ultimately had no post-ablation recurrence in the first year (29% vs. 14%, p = 0.62). In 80% of cannula-adjacent VT, there was a known history of VT prior to LVAD implantation. In all cases of VT termination near the cannula, the morphology was right-bundle type and superiorly directed, with precordial transition between V3 and V5. Cannula-adjacent VT presented a median of 13 days after LVAD implantation, compared with 26 days when related to more remote endocardial scar and 13 days with BBRT (p = NS).
The median follow-up was 269 days (IQR: 160–568), with 33% of patients followed beyond 1 year post-ablation. Eight patients (38%) had recurrence of any VA after ablation, 7 within the first year and 1 at 407 days post-ablation. For 1 patient, post-ablation events were attributed to mechanical irritation (“suction”) from the LVAD inflow cannula, successfully treated with optimization of volume status and LVAD pump speed. One patient developed rapidly conducting atrial fibrillation that induced VT, with acceleration by antitachycardia pacing therapy followed by an ICD shock that terminated both atrial and ventricular arrhythmias. Of the 6 remaining patients with VA recurrence, 3 had a history of VT prior to LVAD, and 4 presented with post-LVAD VT within 30 days after LVAD implantation. Within the first 6 months post-ablation, 4 of 21 patients experienced ICD shocks (19%); only 1 of those patients survived to 6-month follow-up and unfortunately died on day 229 after ablation, just over 2 months after pump exchange for VAD thrombosis.
There was significant reduction in amiodarone use, with 6 of 19 patients who survived to hospital discharge after ablation kept on amiodarone therapy, compared with 17 of 21 patients on amiodarone in the month preceding ablation. Four of those 6 patients discharged on amiodarone survived to 6 months, by which time all but 1 had the dose reduced from 400 mg to 200 mg daily. Two patients not discharged on amiodarone after ablation were started on the drug by 6 months post-procedure, both primarily to control atrial arrhythmias.
Mortality post-ablation is shown in Figure 4. Overall mortality at 6 months was 29% and at 1 year was 47%. Mortality at 1 year among patients without VA recurrence or ICD shocks post-ablation was 33%, compared with 71% for patients with VA recurrence after ablation (p = 0.049). No patient who had ICD shocks after ablation survived to 1 year (p = 0.002 compared with the cohort without either recurrence or ICD shocks). Of note, post-ablation ICD programming was individualized based in part on procedural findings, including cycle lengths of induced VT and ability to terminate with overdrive pacing. However, VT therapy zones were generally set high (typically >190 to 200 beats/min for any shock delivery), with long detection times and multiple (5 to 10) antitachycardia pacing attempts prior to shock delivery. Therapy for VF began at 240 to 250 beats/min. There was a trend toward higher post-ablation survival for patients who presented with VA more than 30 days after LVAD implantation, but these differences did not reach statistical significance. Survival free of any VA also trended higher for patients who presented with late VA, particularly for patients with a mechanism other than BBRT.
In no case was death directly attributed to VA, although recurrent VT (and perhaps ICD shocks) are likely to have contributed to progressive multiorgan failure in the setting of concurrent RV failure in patients 1 and 13; both patients initially presented with VT storm within days of LVAD implantation, and both died within 8 weeks of the ablation procedure. For 3 other patients with VA recurrence (Patients #6, #15, #17), pump dysfunction related to thrombosis and/or hemolysis was implicated. The remaining 4 patients who died by 1 year of follow-up had no VA recurrence post-ablation; infection was a prevalent finding in 3, including 2 who died within 5 weeks of ablation, and 1 (Patient #5) developed intracranial hemorrhage after treatment for LVAD thrombosis.
There were no procedural deaths, pericardial effusions, or LVAD malfunctions. One patient developed a groin hematoma that resolved with conservative management. One patient had suspected pump thrombosis approximately 3 weeks after the procedure, but this was not confirmed with pump exchange and analysis. Five patients in the cohort were ultimately diagnosed with confirmed LVAD thrombosis requiring device exchange, 1 in the inflow cannula and the other 4 in the pump itself—3 died within 3 months of diagnosis. All 5 had early VA after LVAD implantation (p = 0.12 for comparison of thrombosis incidence between early and late VA groups). Two occurred in the HeartMate II cohort and 3 in the HVAD cohort (13% vs. 50% thrombosis rate, p = 0.11). Patients who had any ablation performed near the inflow cannula were significantly more likely to develop VAD thrombosis later (57% vs. 8%, p = 0.03). Thrombosis was diagnosed at a median of 273 days after the ablation procedure (IQR: 156 to 530 days); the shortest interval between the ablation procedure and confirmed thrombosis was nearly 5 months, and 1 occurred over 4 years after ablation. Additionally, 80% of patients with confirmed thrombosis had at least 1 time point with subtherapeutic anticoagulation (INR <2.0) prior to ablation, and 100% had at least 1 subtherapeutic INR recorded after ablation.
We present a large single-center cohort of VT ablation in patients exclusively with continuous-flow LVAD, including patients with the HeartWare HVAD. In contrast to other published studies where LVAD often served as a bridge to cardiac transplantation (12,14–16,23,24), the present data are novel in that only destination LVAD patients were included. The major findings of the present study are:
1. Ablation of VT in patients with destination LVAD results in effective suppression of recurrent VA, with reduced ICD therapies and amiodarone use.
2. Cannula-adjacent VT presented earlier than VT associated with scar remote from the cannula, and surrounding apical scar was common in these patients.
3. Freedom from recurrent VT and ICD shocks after ablation was associated with improved survival. Recurrent VT, particularly in the setting of concurrent RV failure, was associated with early mortality.
4. Bundle-branch re-entrant mechanisms were more prevalent than anticipated and portended an excellent prognosis.
Our experience adds further evidence to support the beneficial effects of VT ablation across various LVAD indications, as summarized in Table 4. Freedom from recurrent VT after ablation has been associated with improved survival in large multicenter retrospective studies (25). As post-LVAD VA has been independently associated with increased mortality (9), there is biological plausibility in the observed association between reduction of recurrent VT and improved survival. The present data, although limited by small sample size and retrospective nature, also support escalating risk for mortality in LVAD patients who had ICD shocks, which is consistent with data from SCD-HeFT (Sudden Cardiac Death in Heart Failure Trial) (26).
Role of the apical inflow cannula
Ziv et al. (27) reported that in patients with the older HeartMate LVAD, 7 of 11 available 12-lead VT electrocardiogram tracings (64%) were compatible with an LV apical site of origin. In prior published series of catheter ablation performed after LVAD implantation, an apical origin was found in 29% to 43% of patients when reported (14–16,23,24). Cantillon et al. (12) reported a predominance of total VT related to intrinsic scar versus the inflow cannula (75% vs. 14%), although the number of patients for whom the cannula played a role was not specified. In patients with apical VT, the inflow cannula was often positioned through an area of antero-apical scar (16,23), although in at least 1 case, there was healthy endocardial voltage in the apex but a VT (with a right-bundle, right superior axis) that revolved around the cannula site (14). Similarly, we found most patients with VT that slowed or terminated near the inflow cannula had some degree of apical scarring, and in all cases, the relevant VT morphology was right-bundle type and superiorly directed, with late precordial transition between V3 and V5.
In contrast to the multicenter report by Sacher et al. (23), in which the reported median time from LVAD implantation to cannula-related VT was 38 days versus 8 days for VT from other locations, we found a higher proportion cannula-adjacent VT presenting early after LVAD implantation. The reasons for these differences are unclear, but it is likely that the categorization of remote endocardial scar versus cannula-related scar is not mutually exclusive, because patients frequently have pre-existing apical scar. This assertion is supported by the fact that 80% of the patients with successful ablation adjacent to the cannula had a known history of VT prior to LVAD implantation. However, the implantation procedure itself may create additional substrate or facilitate re-entry, as there was a trend toward earlier presentation of cannula-adjacent post-implantation VT. Recurrent VT after ablation also appeared to occur more commonly in patients without known pre-VAD arrhythmia, again suggesting the possibility of a newly introduced mechanism related to the presence of the LVAD that may not be fully treatable by endocardial catheter ablation.
Procedural barriers to successful catheter ablation near the cannula include inadequate power delivery on or near the sewing ring, possible interference with the 3-dimensional mapping system, and difficulty with epicardial access after LVAD implantation, although a subxiphoid surgical approach has been previously demonstrated with success (28). These issues raise the question of whether empiric ablation at the time of LVAD implantation could reduces the overall burden of post-implantation VA. Open chest epicardial mapping and ablation for recurrent VT during LVAD implantation has been demonstrated to be feasible in both our own unpublished experience and small published series, using both cooled RF energy (29) and cryoablation (30–32). The long-term safety and efficacy of empiric ablation around the inflow cannula site requires further study.
Bundle-branch re-entrant ventricular tachycardia
An unanticipated prevalence of BBRT of 14% was observed in this cohort. All 3 patients had a history of VT prior to LVAD implantation, although no patient had BBRT as suspected or proven mechanism. Two reports of BBRT after LVAD have been noted in the previously published experience (12,24). BBRT represents a mechanism with favorable prognosis after right bundle-branch ablation, and thus its recognition is important clinically as therapeutic options are weighed in this population. This diagnosis needs to be considered in left bundle-branch block morphology VT, and noninvasive clues from ICD overdrive pacing may be helpful for diagnosis (Figure 5).
LVAD pump thrombosis
A concerning signal of increased incidence of late LVAD thrombosis requiring hardware replacement, particularly in patients with the HeartWare HVAD device, was observed in those who developed VT within 30 days of implantation. The earliest diagnosis of thrombosis was 148 days after the ablation procedure, and median time to diagnosis was 273 days, casting uncertainty on any causal relationship. Many patients also had subtherapeutic and/or inconsistent anticoagulation status within targeted range. Although patients with thrombosis were more likely to have had ablation performed near the inflow cannula, those same patients had scar involving the LV apex, with the potential for more stasis and thrombogenicity near the cannula. Despite the higher incidence of late VAD thrombosis, overall arrhythmia-free survival was better for the HVAD group compared with the HeartMate II group (83% vs. 47% at 6 months, p = 0.15). Periprocedural anticoagulation was carefully controlled and confirmed for all patients, but in this retrospective analysis, it is difficult to control for all possible confounders, including surgical technique, medication compliance, time spent with therapeutic INR, overall burden of VA, and burden of atrial arrhythmias—notably, all but 1 patient with confirmed thrombosis had atrial fibrillation, atrial flutter, or both.
In a recent report of 2 patients who underwent surgical epicardial and endocardial cryoablation immediately followed by HeartMate II LVAD implantation, both required pump exchange (at approximately 3 months and 20 days post-implantation, respectively) for thrombosis (31). An analysis of 382 patients who received the HVAD as part of the BTT (Bridge to Transplant) and subsequent CAP (Continued Access Protocol) trials revealed pump thrombus events in 8.1% of the overall cohort, and VA were more common in patients who experienced device thrombosis than in those who did not (38.7% vs. 18.5% of patients, p = 0.02) (33). Of 77 patients with VA, 12 (15.6%) experienced thrombosis, 8 after the VA and 4 <30 days before the VA. Our findings, with an overall rate of confirmed thrombosis of 23.8%, certainly warrant further investigation—the population of patients with a continuous-flow LVAD placed for destination therapy and who also have refractory VA warranting ablative therapy appears to be at even higher than normal risk for device thrombosis. The MOMENTUM 3 IDE Clinical Study demonstrated the superiority of a fully magnetically levitated pump and holds promise for mitigating the increased thrombosis rate seen in this challenging patient population (34).
This is a retrospective, observational study. The sample size remains small due to the limited patient population of interest, and detailed data on VT burden (particularly before ablation) were limited in the records available for review. Causal relationships cannot be proven based on the present data, and results of hypothesis testing must be approached with caution. All procedures were done at a single center, and all but 1 were led by a single operator, with the advantage of technical and observational consistency but reduced generalizability. The decision to proceed was left to the discretion of the managing team, and the threshold for and timing of ablation were therefore variable. All patients were deemed ineligible for cardiac transplantation, which also potentially limits extrapolation of findings outside the population receiving an LVAD as destination therapy. Information on RV function and how it may or may not have contributed to morbidity and mortality associated with VA was limited by variability in timing of RV assessment relative to arrhythmia presentation and ablation. Lastly, ICD programming was not standardized, which may have influenced the incidence of shocks after ablation, although it is the institutional practice to program higher rate thresholds and long detection times, with multiple attempts at antitachycardia pacing.
In this relatively large single-center experience of VT ablation in patients with LVAD as destination therapy, freedom from recurrent VT and ICD shocks was associated with improved 1-year survival. Patients with cannula-adjacent VT presented earlier after LVAD implantation, and bundle-branch re-entry was more prevalent than anticipated. This challenging population remains at risk for pump thrombosis and mortality.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients with refractory ventricular tachycardia after LVAD implant as destination therapy, freedom from recurrent VT and ICD shocks after endocardial catheter ablation is comparable to success rates in other populations of patients with structural heart disease and is associated with improved survival. The procedure is safe in patients with both the HeartMate II LVAD and HeartWare HVAD, although there may be an increased risk of late thrombotic events, with uncertain relationship to the ablation procedure itself.
TRANSLATIONAL OUTLOOK: Prospective, randomized trials are needed to better assess the efficacy, long-term benefits, and safety of endocardial VT ablation in patients with a continuous flow LVAD.
The authors thank Jonathan Grinstein, MD, for providing data on measures of RV function, and Stephanie A. Besser, MS, MA, for her statistical advice.
Dr. Moss has served as a consultant for St. Jude Medical. Dr. Nayak has received speaker’s honoraria from Biosense Webster. Dr. Upadhyay has received research support from Medtronic and Biotronik. Dr. Jeevanandam has served as a consultant for and scientific advisor to Thoratec; and as a scientific advisor to ReliantHeart and HeartWare. Dr. Uriel has received grant support from and has served as a consultant for Abbott and Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- bundle-branch re-entrant ventricular tachycardia
- ejection fraction
- implantable cardioverter-defibrillator
- international normalized ratio
- interquartile range
- left ventricular assist device
- right ventricular
- ventricular arrhythmia
- ventricular tachycardia
- Received February 28, 2017.
- Revision received May 30, 2017.
- Accepted May 31, 2017.
- 2017 American College of Cardiology Foundation
- Aaronson K.D.,
- Slaughter M.S.,
- Miller L.W.,
- et al.,
- for the HVAD Bridge to Transplant ADVANCE Trial Investigators
- Garan A.R.,
- Levin A.P.,
- Topkara V.,
- et al.
- Yoruk A.,
- Sherazi S.,
- Massey H.T.,
- et al.
- Makki N.,
- Mesubi O.,
- Steyers C.,
- Olshansky B.,
- Abraham W.T.
- Aliot E.M.,
- Stevenson W.G.,
- Almendral-Garrote J.M.,
- et al.
- Uriel N.,
- Sayer G.,
- Addetia K.,
- et al.
- Marchlinski F.E.,
- Callans D.J.,
- Gottlieb C.D.,
- Zado E.
- Stevenson W.G.,
- Friedman P.L.,
- Sager P.T.,
- et al.
- Stevenson W.G.,
- Sager P.T.,
- Natterson P.D.,
- Saxon L.A.,
- Middlekauff H.R.,
- Wiener I.
- Jais P.,
- Maury P.,
- Khairy P.,
- et al.
- Arenal A.,
- Glez-Torrecilla E.,
- Ortiz M.,
- et al.
- Sacher F.,
- Reichlin T.,
- Zado E.S.,
- et al.
- Snipelisky D.,
- Reddy Y.N.,
- Manocha K.,
- et al.
- Tung R.,
- Vaseghi M.,
- Frankel D.S.,
- et al.
- Ziv O.,
- Dizon J.,
- Thosani A.,
- Naka Y.,
- Magnano A.R.,
- Garan H.
- Whang W.,
- Patel M.R.,
- Iyer V.,
- et al.
- Patel M.,
- Rojas F.,
- Shabari F.R.,
- et al.
- McIlvennan C.K.,
- Babu A.N.,
- Brieke A.,
- Ambardekar A.V.
- Najjar S.S.,
- Slaughter M.S.,
- Pagani F.D.,
- et al.,
- for the HVAD Bridge to Transplant ADVANCE Trial Investigators
- Mehra M.R.,
- Naka Y.,
- Uriel N.,
- et al.,
- for the MOMENTUM 3 Investigators