Author + information
- Received March 2, 2017
- Revision received June 23, 2017
- Accepted July 6, 2017
- Published online November 15, 2017.
- Mohit K. Turagam, MDa,
- Venkat Vuddanda, MDb,
- Donita Atkins, RN, BSNb,
- Pasquale Santangeli, MD, PhDc,
- David S. Frankel, MDc,
- Roderick Tung, MDd,
- Marmar Vaseghi, MDe,
- William H. Sauer, MDf,
- Wendy Tzou, MDf,
- Nilesh Mathuria, MDg,
- Shiro Nakahara, MDh,
- Timm M. Dickfeld, MDi,
- T. Jared Bunch, MDj,
- Peter Weiss, MDj,
- Luigi Di Biase, MD, PhDk,l,
- Venkat Tholakanahalli, MDm,
- Kairav Vakil, MDm,
- Usha B. Tedrow, MDn,
- William G. Stevenson, MDn,
- Paolo Della Bella, MDo,
- Kalyanam Shivkumar, MD, PhDe,
- Francis E. Marchlinski, MDc,
- David J. Callans, MDc,
- Andrea Natale, MDk,
- Madhu Reddy, MDb and
- Dhanunjaya Lakkireddy, MDb,∗ ()
- aDivision of Cardiovascular Medicine, University of Missouri Hospital and Clinics, Columbia, Missouri
- bDivision of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital & Medical Center, Kansas City, Kansas
- cCardiovascular Division, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania
- dPritzker School of Medicine, University of Chicago Medicine, Chicago, Illinois
- eUCLA Cardiac Arrhythmia Center, UCLA Health System, Los Angeles, California
- fUniversity of Colorado, Aurora, Colorado
- gSt. Luke’s Health System/Texas Heart Institute and University of Texas Health Science Center, Houston, Texas
- hDokkyo Medical University Koshigaya Hospital, Saitama, Japan
- iUniversity of Maryland Medical Center, Baltimore, Maryland
- jIntermountain Heart Institute, Intermountain Medical Center, Murray, Utah
- kTexas Cardiac Arrhythmia Institute, St. David’s Medical Center, Austin, Texas
- lAlbert Einstein College of Medicine at Montefiore Hospital, New York, New York
- mUniversity of Minnesota Medical Center, Minneapolis VA Medical Center, Minneapolis, Minnesota
- nBrigham and Women’s Hospital, Boston, Massachusetts
- oHospital San Raffaele, Milan, Italy
- ↵∗Address for correspondence:
Dr. Dhanunjaya Lakkireddy, Division of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital and Medical Center, 3901 Rainbow Boulevard, MS3006, Kansas City, Kansas 66160.
Objectives To evaluate the clinical outcomes of patients receiving hemodynamic support (HS) during ventricular tacchycardia (VT) ablation.
Background There are limited real-world data evaluating its effect of HS in ablation outcomes.
Methods An analysis of 1,655 patients from the International VT Ablation Center Collaborative group was performed. A total of 105 patients received HS with percutaneous ventricular assist device.
Results Patients in the HS group had lower left ventricular ejection fraction (LVEF), higher New York Heart Association (NYHA) functional class, and more implantable cardioverter-defibrillator (ICD) shocks, VT storm, and antiarrhythmic drug use (all p < 0.05). The HS group also required significantly longer fluoroscopy, procedure, and total lesion time. Acute procedural success (71.8% vs. 73.7%; p = 0.04) was significantly lower and complications (12.5% vs. 6.5%; p = 0.03) and 1-year mortality (34.7% vs. 9.3%; p < 0.001) were significantly higher in the HS group. Multivariate Cox regression analysis demonstrated HS as an independent predictor of mortality (hazard ratio: 5.01; 95% confidence interval: 3.44 to 7.20; p < 0.001). There was no significant difference in VT recurrence between groups. In a subgroup analysis including LVEF ≤20% and NYHA functional class III to IV patients, acute procedural success (74.0% vs. 70.5%; p = 0.8), complications (15.6% vs. 7.8%; p = 0.2), VT recurrence (30.2% vs. 38.1%; p = 0.44), and 1-year mortality (40.0% vs. 28.8%; p = 0.2) were no different between the HS and no-HS groups.
Conclusions Patients requiring HS were sicker with multiple comorbidities and, as expected, had a significantly higher 1-year mortality than did those patients in the no-HS group. In patients with LVEF ≤20% and NYHA functional class III to IV, there was also no significant difference in clinical outcomes when compared with no HS. Further studies are needed to systematically evaluate patients undergoing VT ablation receiving HS.
Catheter ablation is currently an effective strategy in the treatment of incessant ventricular tachycardia (VT) and to reduce or prevent recurrent, symptomatic, drug-refractory episodes of sustained VT, electrical storm, and implantable cardioverter-defibrillator (ICD) shocks in structural heart disease (1–8). A comprehensive VT ablation procedure involves extensive activation, entrainment, pace, and substrate mapping to localize the isthmus of the circuit to improve success and minimize complications.
However, about 50% to 80% of patients with structural heart disease referred for VT ablation are hemodynamically unstable (2,9). Furthermore, the use of anesthesia, inductions of clinical VT, and volume overload can lead to sustained hemodynamic collapse, decompensated heart failure, and irreversible end-organ damage. However, pace and substrate mapping can be performed without induction of clinically unstable VT it is associated with its share of limitations (10–12).
Recently, a few studies have demonstrated feasibility and safety of using hemodynamic support devices (HS) such as intra-aortic balloon pump (IABP), percutaneous ventricular assist devices such as Impella 2.5 (Abiomed, Danvers, Massachusetts), extracorporeal membrane oxygenation (ECMO), and TandemHeart (TandemLife, Pittsburgh, Pennsylvania) during VT ablation in hemodynamically unstable patients (13–18). However, these studies are limited by small sample size with significant heterogeneity among patient populations and ablation techniques. There are limited data regarding prevalence, predictors and clinical outcomes of patients receiving HS. In this study, we retrospectively examined clinical predictors of long-term outcomes of patients undergoing VT ablation receiving HS and its impact on mortality and VT recurrence compared with those not receiving HS from a large real-world International VT Ablation Center Collaborative (IVTCC) group database.
We performed a multicenter, retrospective study of 1,655 patients from the shared database of the IVTCC group, which comprises 12 specialized arrhythmia management centers worldwide, who underwent VT ablation between 2002 and 2015. VT ablation was performed for standard indication in all patients such as recurrent ICD shocks or VT storm despite antiarrhythmic therapy. VT ablation was performed in 105 patients receiving HS. HS was defined as the use of Impella 2.5 ventricular support device, ECMO and TandemHeart. We excluded patients who received IABP from the HS group. Details of the shared IVTCC database have been previously reported (7). The registry was approved by the Institutional Review Board of each participating specialized center.
Ablation and procedural data collection
Electroanatomic mapping was performed using CARTO (Biosense Webster, Diamond Bar, California) or NavX (St. Jude Medical, St. Paul, Minnesota) systems followed by entrainment, activation, and substrate mapping based on feasibility and clinical judgment of the electrophysiologist. Extent of VT ablation and the use of epicardial access were at the discretion of the operator. Success of VT ablation was defined as complete termination of monomorphic VT and subsequent VT noninducibility despite programmed electrical stimulation unless prohibited due to hemodynamic instability. The use of HS with percutaneous assisted devices such as Impella 2.5 ventricular support device, ECMO, and TandemHeart was also at the discretion of the operator. All HS was placed in the electrophysiology laboratory by the operating electrophysiologist or an interventional cardiologist using standard techniques either before placement of diagnostic catheters or after VT induction depending on the clinical condition. The details of the VT ablation procedure and methods of data collection have been previously published (7). Briefly right and left femoral veins were accessed and venous sheaths were placed. Right ventricular quadripolar catheter was used for right ventricle pacing and induction and overdrive pacing. Access to the left heart was obtained by either transseptal or retroaortic approach. All patients had either radial arterial or femoral arterial lines placed for close monitoring of intra-arterial pressure. Procedures were performed under conscious sedation or general anesthesia.
Patients on oral anticoagulants before the procedure were managed by stopping warfarin 2 to 4 days prior or reversed with fresh frozen plasma for target international normalized ratio of <1.5. novel oral anticoagulant were stopped the night before the procedure and restarted the day after the procedure. Intraprocedurally, unfractionated heparin to keep ACTs 250 s with heparin at 50 to 100 units/kg bolus and 1,000 to 1,500 units/h infusion was administered just before or immediately after the transseptal puncture with TandemHeart or placement of the arterial sheath with Impella 2.5. The arteriotomy sites were closed using a standard technique such as closure devices and manual compression.
In patients in whom epicardial access was pursued after endocardial ablation HS devices were either removed or withdrawn into the descending aorta (if Impella 2.5) followed by reversal of anticoagulation. However, in patients who underwent epicardial access at the beginning of the procedure, HS devices and anticoagulation was withheld until there was no evidence of pericardial bleeding.
Data collection and follow-up
Data were collected prospectively as a part of the IVTCC registry. All subjects were followed post VT ablation with office visits and device interrogations in the outpatient setting (7). Baseline and procedural characteristics as well as clinical outcomes including acute procedural success, procedural complications, hospital deaths, VT recurrence, and 1-year mortality were compared between groups. Major complications such as pericardial tamponade or effusion requiring drainage as well as vascular complications requiring intervention, stroke, and intraprocedural death were recorded. Post-procedural antiarrhythmic and antithrombotic choice was at the discretion of the operator.
Categorical variables are presented as count or percentage and continuous variables as mean ± SD or median (interquartile range). Categorical variables are compared using chi-square test or Fisher’s exact test as appropriate. The data have a hierarchical (or multilevel) structure, with patients clustered within the hospitals. Mixed effect Cox regression models were used to model each outcome as functions of demographic, and medical characteristics to identify the predictors of long-term mortality and long term VT recurrence. Hospital was included as a random effect to estimate the amount of the residual variance between the centers. To identify variables for multiple regression models we fitted a univariate model for each covariate, and identified the predictors significant at a p value < 0.10. Kaplan-Meier survival curves and log-rank test were used to estimate and compare survival differences at 1 year between subjects treated with HS versus those treated without HS.
In patients who received HS, mixed-effects analyses of covariance models were used to model each outcome as functions of demographic, and medical characteristics to identify the predictors of long-term mortality and VT recurrence. Hospital was included as a random effect to estimate the amount of the residual variance between the centers. All results are presented as odds ratios (95% confidence interval). Subgroup analysis was performed in the sickest group of patients with a left ventricular ejection fraction (LVEF) ≤20% and NYHA functional class III to IV. Statistical analyses were performed using SPSS version 23.0 (IBM Corporation, Armonk, New York) and R 3.41 (R Foundation for Statistical Computing, Vienna, Austria). The Cox proportional hazards model was estimated by the coxPH function in the survival package, and the frailty Cox model by the coxme function in the Coxme package. The glmer function in the lme4 package was used for mixed-effects analyses of covariance models.
A total of 1,655 patients were referred for catheter ablation for VT. Of these, 105 (6.3%) patients underwent VT ablation while receiving HS. The mean age of patients in HS group was 63.6 ± 11.2 years, with 85.7% of subjects being men. The mean LVEF was significantly lower (25.4 ± 11.9% vs. 35.6 ± 12.9%; p < 0.001) and there were more patients with NYHA functional class III (53.8% vs. 30.0%; p < 0.001) and NYHA functional class IV (22.1% vs. 3.9%; p < 0.001) symptoms in the HS group versus the no-HS group. Other characteristics are presented in Table 1. Comorbidities such as atrial fibrillation (AF), hypertension, diabetes mellitus, chronic kidney disease, and hyperlipidemia were significantly more frequent in patients requiring HS than in those without HS. There were also significantly more patients with VT electrical storm, ICD shocks, and use of antiarrhythmic drugs including amiodarone in the HS group versus the no-HS group. We calculated the PAINESD (P-pulmonary/chronic obstructive airway disease, age >60 years, ischemic cardiomyopathy, anesthesia, NYHA functional class ≥III, VT storm, and diabetes) score associated with acute hemodynamic collapse during VT ablation (19) without pulmonary and anesthesia components, as they are not available in the IVTCC database. Despite the absence of these components, the PAINESD score was significantly higher in HS versus no HS groups (15.3 ± 6.2 vs. 10.1 ± 6.3; p < 0.0001).
Table 1 describes the procedural characteristics of the patients. There were no significant differences in the proportion of patients undergoing epicardial access, both endocardial and epicardial VT ablation, and mapping between groups. However, the HS group had a significantly higher number of VTs induced and mappable VTs, longer fluoroscopy exposure, total lesion, and procedure times than did the no-HS group.
The median follow-up period in the entire cohort was 527 days (range 208 to 1,048 days). Table 2 compares the clinical outcomes between the HS and no-HS groups. Acute procedural success as defined by the noninducibility of clinical VT was achieved in 71.8% versus 73.7% patients in the HS group versus the no-HS group (p = 0.04). Procedural complications included pericardial tamponade or effusion requiring drainage, vascular complications such as access site hematoma, arteriovenous fistula requiring intervention, systemic thromboembolism, and in-hospital mortality. There were significantly more complications (12.5% vs. 6.5%; p = 0.03), especially pericardial tamponade (5.0% vs. 1.8%; p = 0.02) and in-hospital mortality (21.0% vs. 2.2%; p < 0.001), in the HS group. One-year mortality was significantly greater (34.7% vs. 9.3%) in the HS group versus the no-HS group.
Univariate analysis of predictors of mortality at 1-year follow-up in the HS group is presented in Table 3. Baseline or procedural characteristics including age, ICM, type of device, NYHA functional class, electrical storm, ICD shocks, or type of ablation were not significantly different in those who died versus alive in the HS group. Pre-ablation LVEF (21.7 ± 9.1% vs. 26 ± 12.7%; p = 0.09) was significantly lower, number of VTs induced (2.80 ± 1.94 vs. 2.01 ± 1.24; p = 0.039) was significantly higher, and fluoroscopy time (79.4 ± 27.2 min vs. 57.6 ± 31.5 min; p = 0.036) was significantly longer in those dead versus alive. However, when the 3 variables were included in a multivariate logistic regression model there were not statistically significant (p = 0.17). Cox regression analysis demonstrated HS as an independent predictor of mortality (hazard ratio: 5.01; 95% confidence interval: 3.44 to 7.20; p < 0.001). Kaplan-Meier survival curves show a significantly higher mortality in patients undergoing VT ablation on HS versus those without (p < 0.0001) (Figure 1).
There was no significant difference in VT recurrence between HS and no-HS groups at 12 months follow-up (29.6% vs. 25.4%; p = 0.4). Univariate analysis predictors of VT recurrence at 1-year follow-up in the HS group are presented in Table 4. Significantly higher mean number of VTs induced (3.05 ± 1.76 vs. 2.02 ± 1.35; p = 0.006) and longer procedure time (342.0 ± 134.0 min vs. 275.0 ± 109.1 min; p = 0.04) were seen in HS patients with VT recurrence versus those without VT recurrence. However, these associations were not significant in a multivariate logistic regression model (p = 0.33).
Subgroup-analysis in LVEF ≤20% and NYHA functional class III to IV
Baseline and procedural characteristics
Table 5 describes the baseline and procedural characteristics of patients with an LVEF ≤20% and NYHA functional class III to IV who underwent VT ablation with and without HS. The HS group had significantly more patients with a history of AF (45.7% vs. 24.1%; p = 0.009), diabetes (61% vs. 26.4%; p < 0.001), and hypertension (85% vs. 62%; p = 0.008) when compared with no-HS patients. HS group also had significantly more mappable VTs (64% vs. 41%; p = 0.03) compared with the no-HS group. There was no significant difference in fluoroscopy, procedure, and total lesion time between both groups. The calculated PAINESD score (without pulmonary and anesthesia components) did not significantly differ between the HS and no-HS groups (19.6 ± 4.7 vs. 18.6 ± 4.7; p = 0.2).
Table 6 compares the clinical outcomes including complications and mortality between HS and no-HS patients in this low LVEF subgroup. Acute procedural success was 74% vs. 70.5% in patients with HS versus no HS (p = 0.86). There was lower VT recurrence (30.2% vs. 38.1%; p = 0.44) and higher complications (15.6% vs. 7.8%; p = 0.2), including in-hospital mortality (19.6% vs. 10.6%; p = 0.18) and total mortality (40% vs. 28.8%; p = 0.22), between the groups that did not reach statistical significance.
Univariate analysis predictors of hospital and 1-year mortality in patients with LVEF ≤20% and NYHA functional class III to IV are listed in Online Tables 1 and 2. Kaplan-Meier estimates demonstrate no significant difference in mortality in patients undergoing VT ablation on HS versus those without (p < 0.1) in this subgroup (Figure 2).
Univariate analysis of predictors of VT recurrence at 1-year follow-up in patients with LVEF ≤20% and NYHA functional class III to IV is presented in Online Table 3. Kaplan-Meier estimates demonstrate no significant difference in VT-free survival between patients undergoing VT ablation on HS versus those patients without VT ablation (p < 0.2) in this subgroup (Figure 3).
In this multicenter, observational real-world data from experienced centers, HS was used in 6.3% of VT ablation procedures, in a group of patients who had worse ventricular function and functional class than in the overall population. When compared with no HS: 1) HS patients had significantly lower acute procedural success, higher risk of complications, higher rates of in-hospital mortality and 1-year mortality, but no difference in VT recurrence; 2) HS was an independent predictor of mortality on multivariate Cox regression analysis; and 3) in a subgroup of patients with severely depressed LVEF (≤20%) and NYHA functional class III and IV heart failure receiving HS, there were also no significant differences in acute procedural success, VT recurrence, and hospital and 1-year mortality.
Patients with scar-related VT referred for catheter ablation are complicated and pose a unique challenge due to severe congestive heart failure, episodes of VT recurrence resulting in frequent ICD shocks, electrical storm, and presence of significant comorbid conditions including diabetes, chronic kidney disease, and AF. Hence, careful patient selection and procedural planning is very important to improve efficacy and safety of the procedure. Recently, in patients with structural heart disease there has also been a surge in unstable VTs with shorter cycle lengths due to a change in scar size, burden, and properties due to the widely accepted practice of early coronary revascularization (19). These episodes of clinically unstable VT can lead to hemodynamic decompensation and difficulty in performing adequate activation and entrainment mapping where targeted focal ablation can terminate the arrhythmia (2,9). Furthermore, general anesthesia during the procedure and repetitive induction of VT to assess for noninducibility as a procedural endpoint can lead to hemodynamic compromise and organ hypoperfusion further limiting additional mapping and ablation (20,21). To overcome some of the previous limitations, substrate based-ablation strategies have evolved recently, where ablation can be performed in sinus rhythm without exposing the patient to risks of unstable VT (10,22). However, there are some limitations with this approach such as: 1) substrate-based ablation requires extensive scar homogenization, which may prolong procedure time, increasing risk of complications; 2) low therapy effectiveness in nonischemic cardiomyopathy patients due to difficulty in identifying fractionated and late potentials; and 3) anatomical considerations, especially with clinical VT arising close to critical structures such as the coronary arteries, conduction system, and phrenic nerve. Despite these limitations there is lack of robust data from randomized clinical trials and it remains uncertain if ablation of clinical VT is superior to substrate ablation strategy in this extremely sick group of patients with severely depressed LVEF and high NYHA functional class.
Recently several centers have started to use HS devices such as IABP, percutaneous ventricular assist devices such as the Impella 2.5 system, ECMO, and TandemHeart to assist VT ablation. There are no randomized clinical data comparing these support systems. Among, the various HS devices available, it appears IABP can be efficacious in augmenting diastolic blood pressure in cardiogenic shock when in sinus rhythm but has limited benefit during sustained VT. IABP has an advantage of the ease of placement, smaller vascular access requirements, and greater overall familiarity with the device (13,15,18). ECMO, which remains limited to highly specialized centers, can theoretically provide the greatest HS (up to 5 l/min) even during fast VT and in patients with severe right ventricular dysfunction. ECMO has a disadvantage of being a more invasive strategy, conveys a risk of thromboembolism, and enlarges left ventricular volumes with higher support, which can sometimes lead to difficulty in mapping and ablation of VT (23). However, the bulk of evidence regarding use of mechanical circulatory devices during VT ablation comes with Impella and TandemHeart (13,15,17). Reddy et al. (13) published the largest study to date with HS devices during VT ablation including 66 patients with a mean follow-up of 12 ± 5 months comparing IABP, Impella, and TandemHeart. Although, these devices provide different levels of HS, previous studies have consistently shown that despite greater ability to perform VT mapping and ablation, there were no significant differences in acute procedural success, mortality, or VT recurrence (13–17,24).
The results of our study are consistent with previously published data that patients who receive HS during VT ablation are sicker with severely reduced LVEF, higher NYHA functional class, and multiple comorbid conditions with increased mortality during follow-up (7,25). The subgroup analysis demonstrated that HS group had higher complications (15.6% vs. 7.8%; p = 0.2), including in-hospital mortality (19.6% vs. 10.6%; p = 0.18) and total mortality (40.0% vs. 28.8%; p = 0.22), than did no HS that did not reach statistical significance. Furthermore, multivariate analysis demonstrated HS as an independent predictor of mortality (hazard ratio: 5.01; 95% confidence interval: 3.44 to 7.20; p < 0.001) in our study. The lack of statistical significance can be partly explained by the smaller sample size in the subgroup. A recent series by Santangeli et al. (25) including 193 patients reported 11% of patients with acute hemodynamic decompensation during VT ablation that required HS device or cancellation of procedure with an increased risk of 1-year mortality (38% vs. 7%; p < 0.001). The predictors of hemodynamic collapse in this study were combined into the PAINESD score. Excluding general anesthesia, a score of 9 to 14 was associated with a 6% riskof hemodynamic collapse and the risk increased to 24% with a score of >15 (25). There was a significant difference in the mean PAINESD score (excluding pulmonary) in the HS group versus no HS (15.3 ± 6.2 vs. 10.1 ± 6.3; p < 0.001) in our study, demonstrating these patients were indeed high risk for hemodynamic collapse during the procedure and subsequently had higher mortality. One could argue that ideally, these patients would have benefited from preemptive HS for cardiovascular stabilization, rather than as a “rescue” tool for hemodynamic decompensation anticipated during the VT ablation. A recent study by Mathuria et al. (26) demonstrated that patients receiving rescue percutaneous left ventricular assist device had a significantly higher 30-day mortality compared with a pre-emptive group (58% vs. 3%; p = 0.001), highlighting the need for early identification of high-risk patients and clinical variables that could predict the need for HS (26).
The complication rate associated with HS in our study was 12.5%, which is consistent with prior data with rates ranging from 10% to 14% during VT ablation (13,18). The high complication rate is likely due to a very sick group of patients with multiple comorbid conditions, there remains a small possibility that implantation of HS devices itself can be associated with risk of higher complications, especially major bleeding and pericardial effusions, due to need of therapeutic anticoagulation with HS devices.
Furthermore, in our subgroup analysis of patients with severely depressed LVEF (≤20%) and NYHA functional class III to IV, whom we may expect to receive the greatest benefit from HS, there were no significant improvements in acute procedural success as well as in-hospital and 1-year mortality (40.0% vs. 28.8%; p = 0.22) with HS. The increased mortality in the absence of increased VT recurrence in the HS group is likely attributed to progressive heart failure and hemodynamic decompensation in this sick group of patients with multiple comorbid conditions. Our results also demonstrate that both the entire cohort and in the subgroup analysis HS group had greater mappable VTs including longer fluoroscopy and procedure time. It is possible that HS was a critical aid in mapping those VTs that could have otherwise resulted in hemodynamic decompensation.
Typically, HS devices are used based on the individual physician’s discretion in patients who are relatively sick with a poor LVEF or hemodynamically unstable VT with a hope to improve clinical outcomes by improving cardiac output and vital organ perfusion. Similar to the previous studies, we found that HS devices allow for extensive ablation, mapping more VTs, but this procedure benefit does not translate into improved short- or long-term success rates of VT recurrence or mortality. However, one has to realize the wide heterogeneity in the choice of patients and type of HS devices used, small sample size and VT ablation being a palliative procedure in those who are extremely sick with very high morbidity index. HS remains a very important tool in the management of critically ill VT patients. It provides the necessary support during and after the procedure in a lot of these sick patients. There is a significant selection bias against these devices as they are often used in the sickest of the VT ablation group. What we lack is data to support the type of patient groups where HS can offer significant cost benefit ratio. Until a prospective randomized controlled trial is done to systematically identify the appropriate patient population who would gain the most benefit, operators should evaluate the risks, benefits, and costs based on patient and procedural variables we should judiciously use this very important technology during VT ablations.
The study is a retrospective nonrandomized cohort study and is associated with inherent potential bias with unmeasured confounding variables that must be considered during result interpretation. Selection bias regarding which patient and type of HS used during ablation was at the discretion of the operator. The complications reported in the HS group can be variable based on type of HS used and access route. Although majority of the patients received Impella 2.5, followed by TandemHeart we are unable to separate individual patients who received each individual HS device and if HS was pre-emptive or as a rescue. We do not have data regarding rescue defibrillation shocks received during the procedure between groups nor do we have data on antiarrhythmics used post–VT ablation. As noted, data from the IVTCC group represent those of high-volume, experienced centers for VT ablation, and therefore the findings may not be generalizable to all centers performing VT ablation. The number of VTs ablated is fewer than some of the previous published data and this is attributed to severe cardiac dysfunction, sicker patients, and greater hemodynamic instability during procedures. We do not have data regarding cerebral oximetry, which could provide insight regarding extent of HS provided. Despite these limitations, this represents the largest observational study to date investigating the use of HS in VT ablation and provides insight into the feasibility and safety of this practice, as well as its important impact on outcomes.
Patients receiving HS were sicker with multiple comorbidities and as expected, had a significantly higher 1-year mortality than did those not receiving HS. In patients with LVEF ≤20% and NYHA functional class III to IV, there was also no significant difference in clinical outcomes between the 2 groups. Further prospective studies are needed to systematically evaluate patients undergoing VT ablation receiving HS.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients receiving or needing HS during VT ablation have high-risk features of mortality. During selection of HS devices during VT ablation, operators should evaluate the risks, benefits, and costs based on patient and procedural variables.
TRANSLATIONAL OUTLOOK: Given the implications for both higher gain but also higher cost; a prospective comparison of the use of HS in predefined high-risk features should be undertaken.
All the authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- extracorporeal membrane oxygenation
- hemodynamic support
- intra-aortic balloon pump
- implantable cardioverter-defibrillator
- ischemic cardiomyopathy
- International Ventricular Tachycardia Ablation Center Collaborative
- left ventricular ejection fraction
- New York Heart Association
- ventricular tachycardia
- Received March 2, 2017.
- Revision received June 23, 2017.
- Accepted July 6, 2017.
- 2017 American College of Cardiology Foundation
- Calkins H.,
- Epstein A.,
- Packer D.,
- et al.
- Carbucicchio C.,
- Santamaria M.,
- Trevisi N.,
- et al.
- Stevenson W.G.,
- Wilber D.J.,
- Natale A.,
- et al.
- Tung R.,
- Vaseghi M.,
- Frankel D.S.,
- et al.
- Sapp J.L.,
- Wells G.A.,
- Parkash R.,
- et al.
- Soejima K.,
- Suzuki M.,
- Maisel W.H.,
- et al.
- Di Biase L.,
- Burkhardt J.D.,
- Lakkireddy D.,
- et al.
- Fernandez-Armenta J.,
- Penela D.,
- Acosta J.,
- et al.
- Brunckhorst C.B.,
- Delacretaz E.,
- Soejima K.,
- Maisel W.H.,
- Friedman P.L.,
- Stevenson W.G.
- Reddy Y.M.,
- Chinitz L.,
- Mansour M.,
- et al.
- Aryana A.,
- Gearoid O'Neill P.,
- Gregory D.,
- et al.
- Miller M.A.,
- Dukkipati S.R.,
- Mittnacht A.J.,
- et al.
- Miller M.A.,
- Dukkipati S.R.,
- Chinitz J.S.,
- et al.
- Josephson M.E.,
- Anter E.
- Yokokawa M.,
- Kim H.M.,
- Baser K.,
- et al.
- Silberbauer J.,
- Oloriz T.,
- Maccabelli G.,
- et al.
- Nakahara S.,
- Tung R.,
- Ramirez R.J.,
- et al.
- Miller M.A.,
- Dukkipati S.R.,
- Chinitz J.S.,
- et al.
- Santangeli P.,
- Muser D.,
- Zado E.S.,
- et al.
- Mathuria N.,
- Wu G.,
- Rojas-Delgado F.,
- et al.