Ablation of Refractory Ventricular Tachycardia (VT) Using Intra-Myocardial Needle Delivered Heated Saline-Enhanced Radiofrequency (SERF) Energy: A First-in-Man Feasibility Trial

Abstract

Background -

Ablation of VT is limited by the inability to create penetrating lesions to reach intra-myocardial origins. Intramural needle ablation using in-catheter, heated Saline Enhanced Radio Frequency (SERF) energy uses convective heating to increase heat transfer and produce deeper, controllable lesions at intramural targets. This first-in-human trial was designed to evaluate the safety and efficacy of SERF needle ablation in patients with refractory VT.

Methods -

Thirty-two subjects from 6 centers underwent needle electrode ablation. Each had recurrent drug-refractory monomorphic VT after ICD implantation and prior standard ablation. During the SERF study procedure, one or more VTs were induced and mapped. The SERF needle catheter was used to create intramural lesions at targeted VT site(s). Acute procedural success was defined as non-inducibility of the clinical VT after the procedure. Patients underwent follow-up at 30 days, and 3 and 6 months, with ICD interrogation at follow-up to determine VT recurrence.

Results -

These refractory VT patients (91% male, 66±10 yrs., EF 35±11%; 56% ischemic and 44% non-ischemic) had a median of 45 device therapies (shock/ATP) for VT in the 3–6 months pre-SERF ablation. The study catheter was used to deliver an average of 10±5 lesions per case, with an average of 430±295 sec of RF time, 122±65 min of catheter use time, and a procedural duration of 4.3±1.3 hours. Acute procedural success was 97% for eliminating the clinical VT. At average follow-up of 5 months (n=32), device therapies were reduced by 89%. Complications included 2 periprocedural deaths: an embolic mesenteric infarct and cardiogenic shock, 2 mild strokes, and a pericardial effusion treated with pericardiocentesis (n=1).

Conclusions -

Intramural heated saline needle ablation showed complete acute and satisfactory mid-term control of difficult VTs failing 1–5 prior ablations and drug therapy. Further study is warranted to define safety and longer-term efficacy.

Graphical Abstract

Introduction

Ablation is an important component of the therapeutic armamentarium for treating life threatening ventricular tachycardias (VT) in patients with heart disease^1–5. Over the past several decades, a variety of different techniques and technologies have been utilized to facilitate more effective ablation^6–14. Although advances including changes in ablation and electroanatomic mapping techniques, and methods for defining the underlying arrhythmia substrate have improved acute and long-term success rates^{15, 16}. outcomes with standard radiofrequency ablation continue to be disappointing^{3, 4, 6, 7, 10, 11, 13, 17}. In multicenter trials, approximately 50 to 80% of patients treated with ablation for post infarction VT experience recurrences, and outcomes are somewhat more limited for those with nonischemic cardiomyopathy^18–21.

Origin of the arrhythmia substrate deeper to the endocardium, beyond the reach of conventional RF ablation is an important cause of ablation failure^{22, 23}. A recent human study observed that some or all of the arrhythmia substrate is intramural in the majority of scar-related VTs²⁴. As a result, there has been intensified focus on identifying innovative approaches to ablation including the revisiting of early DC Pulse delivery²⁵, as well as the use of photon and particle therapies^{26, 27–30}.

Heated saline-enhanced radiofrequency (SERF) ablation has also been forwarded as a potentially more effective means of creating larger, more penetrating lesions^{26, 30–33}. This approach incorporates a heating stage within the catheter tip so that heated saline can be injected though the needle directly into the target tissue as RF current is delivered. Heated saline injection markedly increases convective heat transport during RF application. Lesion size increases as the duration of RF application is increased and can be sufficient to create transmural left ventricular lesions from an endocardial catheter position^{27, 29, 34, 35}.

The purpose of this study was to evaluate the technical feasibility and procedural safety of using a novel needle electrode catheter that disperses heated saline into the tissue combined with RF energy applied from the needle to treat recurrent sustained, monomorphic VTs that have been refractory to antiarrhythmic drug therapy, ICDs, and conventional catheter ablation therapy. These results were presented as a Late Breaking Clinical Trial at the Heart Rhythm Society 2020 meeting²⁸.

Methods

The study was designed as a non-randomized, open-label, single group, interventional study, conducted to evaluate the safety and performance of the Durablate Catheter and Thermedical Ablation System (Thermedical, Waltham, MA) in subjects with VT. The studies were separately approved under US IDE and Canadian Clinical Trial Authorization and by all sites Institutional Review Boards. All patients provided informed consent. The data methods used in the analysis, and materials used to conduct the research are available for reproducing the results or replicating the procedure.

Patient Characteristics

Thirty-two patients with refractory VT due to structural heart disease, each of whom had previously undergone defibrillator (ICD) implantation and a subsequent standard catheter ablation were enrolled between January 2017 and March 2020. Inclusion and exclusion criteria are listed in Table 1. All patients had been treated with one or more antiarrhythmic agents and an ICD and had recurrence of VT despite treatment with one or more post-ICD implant standard RF ablations.

Table 1:

Inclusion and Exclusion Criteria for the SERF Needle Ablation Trial

INCLUSION CRITERIA
Subject has sustained, monomorphic VT Including recurrent, symptomatic VT Subject has drug refractory or drug intolerant sustained VT following use of at least one Class III antiarrhythmic ECG and/or ICD evidence of a spontaneous VT recurrence within the 6 months after standard ablation A minimum 3-month ICD interrogation history available for evaluation LVEF > 20%, confirmed by echo or comparable technique At least 18 years old Subject is willing and able to provide informed consent and participate in all study procedures
EXCLUSION CRITERIA
VT of idiopathic origin VT with ECG or MRI/CT findings suggestive of right ventricular free wall origin Myocardial infarction (MI) or unstable angina within previous 60 days Subject with cardiac surgery or percutaneous coronary intervention (PCI) within previous 60 days Class IV (NYHA) heart failure Prosthetic cardiac valve(s), severe aortic stenosis or flail mitral valve LV assist device planned or required for the procedure Co‐morbidities with life expectancy less than 1 year Significant intracardiac and/or laminated thrombus evident by transesophageal echo (TEE) or transthoracic echo (TTE) (with contrast if indicated) prior to ablation procedure Thrombocytopenia or another coagulopathy Women who are or may potentially be pregnant (must be post‐menopausal or HAVE negative pregnancy test) Other acute illness or active systemic infection Significant congenital heart disease or cardiac anomaly Allergy or contraindications to the medications/agents used during a standard ablation/EP intervention. Subject concurrently enrolled in any other investigational drug or device study that would interfere with study results

Substrate and Arrhythmia Assessment

The target site of the VT was assessed from previously obtained data including: the 12-lead ECG of the VT, ICD interrogation, mapping data from prior procedures, and cardiac imaging with echocardiography, CTA with delayed enhancement, or MRI with gadolinium in attempt to identify areas of potential scar that could contribute to the arrhythmia substrate. Patients were then taken to the clinical electrophysiology laboratory and catheters placed according to the centers’ standard protocols. Intracardiac ultrasound image analysis was undertaken in real time to identify the location of potential scar, and specific ventricular anatomy. The baseline study included programmed stimulation for induction of VT, if VT was not present spontaneously. The protocol required programmed stimulation using two drive cycle lengths from two right ventricular sites with up to three extra stimuli. LV access was via transseptal approach using a deflectable sheath (Agilis, Abbott) or retrograde aortic approach using an SL-1 sheath (Abbott), depending on the anticipated target VT location. Initial mapping during sinus rhythm and / or VT was generally performed with a conventional electrode catheter and electroanatomic mapping system (Carto, Biosense Webster, Inc.; Precision, Abbott, Inc.; Rhythmia, Boston Scientific, Inc.).

Needle Ablation Device

The Thermedical (SERF) Ablation System’s Durablate deflectable catheter (Figure 1), incorporates a 25-guage stainless steel needle inserted into the myocardium to deliver the RF energy in conjunction with injected heated saline through 30 side holes in the needle. Electrograms could be recorded from 4 distal electrodes at the tip of the catheter. The electrosurgical generator provided radiofrequency (RF) energy up to 50 Watts, simultaneously with the delivery of saline heated to 60° C at a flow rate of 10 ml/min (Figure 2)^28–33.

Figure 1.

SERF Needle Electrode and Ablation Generator System. The deflectable catheter (Panel A) contains a needle that can be extended to 4 or 8 mm depths. Panel B show the ablation system console with monitoring utilities. Panel C shows the range of programmable settings for heated needle energy delivery.

Figure 2.

Lesion Creation with Standard RF and SERF Heated Saline Ablations. Panel A shows RF catheter tip/tissue contact and a representative endocardial surface lesion using a standard catheter design. Panel B is a rendition showing the positioning of the SERF catheter along with the needle extension into the myocardium. Heated saline is injected through side holes into the tissue, markedly increasing convective heating (orange arrows) during RF application to the needle. Panel C shows a representative spectrum of needle ablation lesions with size and depth of extension to the mid and epicardial myocardium from a previous animal model. [31, 32] [29] [33] [28–30]

Ablation Approach

Energy was delivered at selected endocardial sites felt to be in close proximity to the VT origin based on imaging, electrograms, and mapping data. A 12-lead ECG or 12-lead Holter or information from a prior ablation with matching follow-up data showing an identical VT was required. This was used where possible, as that replicating the clinical VT. Two channel morphologies from an ICD was not felt to be adequate to call a “clinical VT.”

The ablation catheter tip was positioned in a preferred perpendicular catheter tip/tissue orientation. The extendable needle was deployed into the target ablation area, at an intra-myocardial position confirmed by fluoroscopy or ICE. The needle depth was selected as 6 to 8 mm depending on the total thickness of the myocardium and any observed scar on ICE. Saline was injected through the needle into the myocardium immediately before and continuing during RF application to increase heat transfer capacity and thereby carry thermal energy deeper into the tissue. Subsequent catheter tip/tissue contact, RF deliveries and lesion development were monitored by intracardiac ultrasound and biplane fluoroscopy. Failure to completely insert the catheter needle into the myocardium produced a flurry of microbubbles as saline flowed back into the LV cavity from the needle as easily visualized on ICE. In the presence of significant (grade 2–3) microbubbles^{27, 29, 30, 34–37}, energy delivery was terminated, and the catheter repositioned. Ablation settings were selected based on preclinical studies to achieve a desired lesion volume and depth (Table 2). The acute endpoint of the ablation procedure was the non-inducibility of the clinically relevant VT. In 14 cases, a conventional ablation catheter was also used for residual ablation targets that were not felt to require intramural ablation. An attempt to ablate all inducible VTs was not required but was left to the discretion of the primary site operator.

Table 2:

Ablation Settings

Setting Number	Duration (s)	Power (W)	Flowrate (mL/min)	Saline Temperature (°C)	Target Lesion Volume* (cc)
1	30	20	10	60	0.50
2	40	25	10	60	1.10
3	55	35	10	60	2.30
4	65	45	10	60	3.60
5	80	50	10	60	5.00

S = Seconds; W = power in watts; ml/min = flow rate in milliliters/minutes; °C = degrees in Centigrade; cc = cubic centimeters anticipated with settings 1–5.

^*from references [27, 36]

Study Endpoints

The trial design sought to establish both safety and efficacy information. Primary Safety Measures were defined as serious adverse events within 30 days that were definitely device related. The Primary Acute Efficacy Measure was non-inducibility of the targeted clinical VT at end of ablation procedure. Secondary Safety Measures included: 1) freedom from acute procedural complications probably or definitely related to the ablation procedure (within 24 hours of ablation) 2) 48 hour prolonged duration of hospitalization (days) following ablation; 3) Number of probably or definitely device related anticipated SAEs within 6 months following ablation; 4) Number of hospitalizations for recurrent VT up to 6 months following ablation; and 5) Percentage of subjects experiencing all-cause and cardiovascular mortality within 6 months of follow-up.

Follow Up

Study visit follow-ups occurred at 30 days, and at 3 and 6 months, and included ICD device interrogation, and assessment of incidence of recurrent VT. Specific patient well-being measures, neurologic function and adverse event occurrence at all follow-ups captured parallel information.

Statistical Methods

Continuous variables are provided as mean ± standard deviation along with the range of that variable, and significance of differences computed using paired T or Kruskal-Wallace testing. Categorical comparisons were made with Chi Square or Fischer Exact test methods. The use of parametric versus nonparametric methods was dependent on the uniformity of distribution of the individual parameter values. A p<0.05 was taken as significant.

Results

Demographics

The 32 patients enrolled (Table 3) had a median age of 67 years, 91% were male; 18 (56%) had ischemic heart disease, of whom 72% had a prior myocardial infarction and 38% had undergone a prior CABG. Nonischemic heart disease producing cardiomyopathy were present in 14 (44%) (Table 3). The mean LVEF was 37%, and the majority of patients had class II (50%) or III (22%) heart failure. These patients had received an average of 45 ICD (shock or ATP) therapies in the preceding 3 to 6 months. Of the 32 patients, 59% had 1 prior ablation procedure, 22% had 2, and 19% had 3 – 5 prior ablation procedures, respectively. All had previously received endocardial procedures and 22 % had also undergone epicardial ablation. One of the patients also underwent trans-coronary ethanol ablation and a needle ablation using another investigational catheter.

Table 3:

Characteristics of the 32 Patients Enrolled in the SERF Trial

Patients (n)	32
Age (years)	67± 9
Gender
Male	29 (91%)
Female	3 (9%)
Ischemic Cardiomyopathy	18 (56%)
Prior MI	72%
Prior CABG	36%
Nonischemic-Cardiomyopathy	14 (44%)
Idiopathic	8
Sarcoid	3
Lamin A/C	1
Lymphocytic CM	1
Late stage HCM	1
EF	37%
NYHA Class
Class I	29%
Class II	45%
Class III	26%
VT Subtype
Monomorphic (32)	100%
Storm (n=3)	9%
Incessant	0%
Prior VT Ablation
Endocardialy	100%
Epicardialy	27%
Endo- and Epicardialy	27%
Diabetes Mellitus
Chronic Renal Disease	22%
Peripheral Vascular Disease	25%
Prior stroke (CVA/TIA)	16%
COPD	22%
AF/AFL	28%

n = Number of subjects; MI = myocardial infarction; CABG = Coronary artery bypass grafting; A/C = arrhythmogenic cardiomyopathy; HCM = hypertrophic cardiomyopathy; EF = ejection fraction; NYHC = New York Heart Association; = ; CVA = cerebral vascular accident; TIA = transient ischemic attack; COPD = chronic obstructive pulmonary disease; AF/AFL = atrial fibrillation or atrial flutter

Mapping and Structural Characterization

Pre-procedure imaging was obtained in 25 patients with CTA, including delayed contrast assessment, while 1 underwent MR assessment including late gadolinium enhancement.

Patients had a median of two different inducible VTs. At least one clinical VT was inducible in 88%, and nonclinical VTs were inducible in 45% of patients. The relevant VT was mappable in 49% of patients. Ablation was guided by substrate assessment combined with mapping during VT in the majority of patients (Figures 3 and 4). In 3 patients ablation was guided only by substrate assessment and use of the VT QRS morphology to determine the likely region within extensive scar areas that gave rise to the VT.

Figure 3.

A patient with recurrent VT due to nonischemic cardiomyopathy had 4 prior ablation procedures, with the last procedure using both endocardial and epicardial approaches. Panel A shows a pre-procedural CT registered with an electroanatomic endocardial activation map of VT. In the CT, green areas indicate thin regions of likely scar whose borders define a potential intramural channel (white rectangle). The activation map of VT (red earliest) shows an area of focal endocardial break out (earliest ventricular activation) overlying this channel. Within this region several areas of fractionated activity were present. The orange markers are ablation lesion sites. Following ablation, VT was no longer inducible. B. Endocardial voltage map suggested a large region of scar. Needle lesion sites are identified by the orange markers. Purple indicates electrogram amplitude > 1.5 mV.

Figure 4.

Mapping and Intracardiac Ultrasound Imaging from a patient with prior infero-posterior myocardial infarction and recurrent VT that had failed to be controlled by prior ablation procedures. Panel A shows the activation map of the targeted VT. Panel B shows the site of ablative lesions. Panel C shows a pre-ablation ICE image of the catheter tip/issue positioning and the inserted needle into the underlying myocardium. Panel D show the evolving ablation lesion indicated by the increasing contrast and intensity of the local tissue seen on ICE. Five needle ablation lesions were applied at the endocardial break out region and abolished this inducible VT, which remained absent during 6 month follow-up.

Ablation

Subsequent positioning of the ablation catheter is shown by real time ultrasound in Figure 4. Also seen is the catheter tip and the extended needle. The number of RF deliveries averaged 10 ± 5 per patient. The depth of needle penetration was 4 or 8 mm as shown in Figure 5A. The ablation profile settings are shown in 5B, with locations in 5C. Median total ablation time was 6.5 minutes, with a median saline volume of 91 mls delivered. The SERF catheter was in place an average of 105 minutes and the total procedure time averaged 4 hours and 42 minutes. Standard RF lesions were used in 14 patients.

Figure 5.

Characteristics of Needle Ablation of VT. A shows the number of energy deliveries at each depth. B shows the number of deliveries at each ablation setting seen in Table 2. C shows an LV polar map indicating the anatomic distribution of energy deliveries in all patients.

Efficacy

Acutely, 97% of testable patients were non-inducible for their clinical VT at the end of the procedure, while no repeat programmed stimulation was performed in one patient due to reduced BP. Median follow-up time was 5 months and the impact of ablation of VT is shown in Figure 6. Prior to needle ablation, these patients required up to 720 shocks or ATPs to treat recurrent VT, as shown in Figure 6B. At 6 months follow up after the procedure, recurrent VT occurred in 50% of patients with 78% of patients remaining on an antiarrhythmic drug at the end of follow up. During the 6 months before the needle ablation, there were 33 ± 81 VT episodes and 18 ± 45 after ablation [p = .0003]. 18 patients receiving shocks before and 9 after (p = .029) [Fig. 6]. Of these patients, 2.0 ± 4.4 (range 0–23) received shocks before and 0.9 ± 2.1 (range 0–10) [p = .0075] after the needle ablation. There was a significant difference in ATPs before 31.5 ± 81.5 (range 0–414) versus 17.1 ± 45.3 (range 0–167) after ablation (p = <.026).

Figure 6.

Pre- and Post-Procedure VT Occurrences, and ICD/ ATP Therapies Under Both Circumstances. The blue bars indicate the pre-ablation frequencies of ICD shocks and ATP, while the green bars show the recurrences of the clinical events after ablation. ICD/ATP therapies were totally eliminated in 50% of all patients; there was a more than 90% reduction in 61% and more than 75% reduction in 78% of all patients, respectively. The spread in the standard deviation described on page 11 reflects the observation that several patients had ongoing ATPs, although this was seen in only 3 patients.

Adverse Events

Adverse events are shown in Table 4. Embolic events occurred in 3 patients. One patient suffered a mesenteric embolic infarct that was fatal. Although unproven, this event may have been related to an air embolus via the trans-septal sheath. A second patient reported a visual field cut, which had been present prior to the ablation, but unreported by the patient. There was no evidence of air emboli, and the event was more likely due to a classical embolic CVA event. One patient sustained a minor stroke and recovered without extensive sequelae over 4 months. Another patient developed a femoral pseudo-aneurysm in addition to an increase in a preexisting pericardial effusion from a prior ablation requiring pericardiocentesis of 300cc of blood. Two patients died within 7 days of the ablation, including the patient with the embolic bowel infarction. The other had ischemic cardiomyopathy and developed refractory cardiogenic shock after extensive ablation for multiple VTs, exiting from a large inferolateral to septal infarct scar. Autopsy showed extensive intramural bleeding through the infarct region dissecting into adjacent myocardium. Seven patients required recurrent hospitalization for VT. Overall, six percent of patients had serious adverse events (Table 4), probably or definitely related to the device, all within 30 days of the ablation, while serious adverse events probably or definitely related to the device after 30 days were not observed. Of the serious adverse events possibly related to the device, 83% were seen within the first 14 patients, while only one occurring in the subsequent 18 patients, possibly reflecting a favorable learning curve with the system.

Table 4:

Adverse Events with Needle Electrode Ablation

	Serious Adverse Events
Stroke/TIA	2
Pericardial effusion requiring drainage	1
Bowel infarct - Fatal	1
Cardiogenic shock - Fatal	1
Femoral pseudoaneurysm	1
Recurrent VT requiring hospitalization (n=7)	7
Heart failure requiring hospitalization	1
Death 6 mos. post ablation: Due to chronic heart failure and sepsis	1
Acute asthma exacerbation	1

^*One death occurred after the bowel infarction on day 2 after ablation; one death occurred at day 1 after ablation due to refractory cardiogenic shock and endocardial dissection; one death occurred at 6 months due to heart failure and sepsis.

Discussion

This study presents the initial clinical outcomes of patients undergoing ablation with a novel deployable needle and heated saline-enhanced RF ablation (SERF) system to eliminate VT. It demonstrates that this system can be effective in reducing VT burden in a high-risk group of patients with severe disease and difficult VT that failed to be controlled by prior antiarrhythmic drugs, ICD therapy and conventional catheter ablation. In addition, adverse events were seen to decrease with experience. These results are encouraging, suggesting that SERF ablation has the potential to improve ablation efficacy over conventional RF ablation, and supports continued evaluation of SERF VT ablation in a larger number of patients.

Traditional Ablation for Ventricular Tachycardia

The utility of RF ablation for reducing VT recurrence is well established^{1–3, 23}. VT in patients with ischemic heart disease is eliminated in 50–80% of patients^{1–3, 5, 13, 17, 23, 38–41}. In non-ischemic dilated cardiomyopathy success rates are somewhat lower in the range of 40–75%, ^{1, 5, 18, 42–45}, although they may have higher VT recurrence and all-cause mortality^{40, 46, 47}. Still, the recurrence rate in both ischemic and non-ischemic patients remain significant. VT in patients with structural heart disease is usually due to reentry in regions of scar. Ablation seeks to destroy viable strands of muscle participating in these reentry circuits or channels^{2, 47–50}. Fibrous scar tissue surrounds areas of both surviving and damaged myocytes and creates the reentry substrate, which can be subendocardial, subepicardial, intramural or transmural. Portions of reentry circuits commonly extend deep to the endocardium and are a recognized reason for failure of VT ablation⁵⁰. RF ablation has been the favored approach, in part because of ease of use^{27, 29, 30, 34–36, 49}. Despite these innovations, radiofrequency (RF) ablation has remained the mainstay of ablation, albeit with disappointing success rates for eliminating VT depending on the number and type of VTs^{3, 7}. A wide range of approaches from DC shock delivery, or particle or photon therapy have been more recently developed with the goal of improving ablation outcomes^{25, 51–53}.

SERF-Mediated VT Ablation and Mechanisms

Traditional RF ablation relies on creation of an area of resistive heating immediately adjacent to the electrode, with conduction of that heat into deeper tissue. Heat transfer by thermal conductivity through tissue is an inefficient process and lesion size is limited by the temperature that can be achieved in the area of resistive heating without producing char or steam pops. The SERF catheter heats saline near the tip of the catheter such that heated saline is injected directly into the myocardium through the side holes in the intramural needle during RF delivery. The saline flow is confined between the endo and epicardial borders. The resulting dramatic increase in “in wall” convective heating can markedly increase lesion size. Preclinical studies have shown creation of controllable lesions that can be large and transmural, extending from the endocardial to the epicardial surface, when desired^{27, 29, 30, 34–36, 49}. The anticipated flow of saline follows in a direction along fiber orientation^{27, 29, 30, 34–36, 49}.

By histologic inspection in animal models, needle lesions are highly uniform, with reliable creation of full-thickness lesions covering scar and ventricular epicardium, without creating potentially pro-arrhythmic irregular border zones. The lesions size ranged from 10×10 to 22×22 mms, depending on the power utilized and the duration of the deliveries, with these data used to develop parameters to predict lesion volumes (Table 2)^{30, 34}. The lesion size in this study was identified acutely on intracardiac ultrasound. In the absence of follow-up CT and MR studies, further definition of lesion size was not possible, thereby excluding the specification of the actual volumes achieved with ablation in these patients. These will be evaluated in subsequent components of this study. Preclinical studies have nevertheless shown an 89% likelihood of eliminating scar channels seen on histology^{27, 48, 50}.

Catheter Deployment

Effective lesion creation requires insertion of the needle into the myocardium. It should be noted that catheter placement for lesion creation could be guided by either biplane fluoroscopy or intracardiac ultrasound. No other ancillary technology was specifically required. Nevertheless, the ability to identify a perpendicular catheter tip/tissue orientation at the endocardial surface is thought to be important to ensure direct delivery of the heated saline into the tissue rather than losing it into the blood pool. With incomplete insertion, some of the side pores of the needle can inject saline into the blood pool, producing microbubbles that are readily evident on ultrasound inspection, and providing a further means of assessing appropriate positioning of the needle. The success of the energy delivery was also likely related to the excellent tissue contact facilitated by insertion of the needle into myocardium.

Identification and Elimination of Intramural VT Substrate

The VT origins in these patients are chronicled in Figure 5c in both the LV septum and free walls, with deeper or sub-epicardial locations, as shown by cumulative needle extensions (Figure 5A). The majority of VT cases were related to ischemic heart disease, thought to be present in 75% of patients dying of SCD¹. VT in patients with coronary artery disease can arise in areas of thin tissue in the border zones of the infarction. In some cases, the site of endocardial break out appears to be a reasonable marker for underlying arrhythmia substrate (Figures 3 and 4). Pre-procedure imaging with CT or MRI may also be helpful in identifying intramural channels (Figure 3)^{7, 11, 54}. Intracardiac ultrasound information may also be helpful to suggest a general target region and for monitoring ablation lesion size^{4, 49, 54–56}.

Safety

Although it might be thought that the insertion of a needle into the myocardium would be accompanied by an increased risk of perforation, this preliminary study demonstrated that this occurrence was rare. There was only 1 patient who had a prior pericardial effusion in whom the effusion increased, requiring drainage. A previous trial with a longer needle also observed that needle insertion into the myocardium is relatively safe⁵⁷.

SERF ablation can produce very large lesions^{27, 34, 49}, raising concern for a risk to viable myocardium outside of the VT substrate scar and worsening ventricular function. One patient in this study, who had extensive substrate and received the largest number of RF applications in multiple areas around an extensive infarct scar did develop fatal cardiogenic shock. Areas of dissecting intramural hemorrhage were seen on autopsy. This led to guidance limiting the total SERF ablation volume to be delivered in one procedure. This study shows, however, that the procedure can be done without excessive injury to residual myocardium outside of the scar or target region. Monitoring catheter placement with intracardiac ultrasound and limiting the number of ablation applications to those required for the targeted VT is prudent. As with standard VT ablation it is important to deliver in regions with pre-existing scar when possible. Use of real-time imaging by intracardiac ultrasound and electroanatomic mapping to delineate regions of scarred vs viable myocardium to correctly identify VT substrate should be critical. There were no steam pops or perforations observed in this study, as is consistent with the findings from preclinical investigations that defined the parameters used. Avoidance of residual myocardium as “an organ at risk” remains critical in reducing risk of the procedure. Furthermore, as the operators gained experience with the device, the occurrence of adverse events decreased.

Three embolic events occurred. Two were middle cerebral artery events. One of two strokes likely occurred prior to ablation, based on symptoms, and was concealed by the patient from investigators for fear it would cancel his procedure. One embolic event (bowel infarct) could have been due to air embolization. It remains unclear whether these were related to the ablation itself, or specifically related to the catheter under study. These emphasize the critical importance of anticoagulation through the course of the procedure and the importance of pre-procedural risk of embolic from the left atrial appendage in patients with AF or the ventricle in patients with poor ventricular function. Overall, the rate of complications is consistent with that seen in other studies of conventional and epicardial RF ablation despite the fact that this was a high-risk population who had already failed prior conventional ablation^{8, 19}. Nevertheless, 7 patients required rehospitalization for recurrent VT.

Limitations

This study has limitations by the nature of its feasibility trial design. Subsequent studies of a greater number of patients will be needed to further establish longer term efficacy and safety. The study did not provide data on the correlation of actual lesion sizes to those predicted from preclinical studies as post-procedural CT or MRI scans were not required. While the lesions were seen acutely on intracardiac ultrasound, the final lesion size can only be inferred from prior animal studies^{27, 29, 35–37, 49}.

In 14 cases, endocardial ablation with a standard catheter was also performed, often when additional VTs were encountered that were not felt to require deep ablation lesions. This endocardial ablation could also have had an impact on outcome. Of note, it is possible to have shallow ablation requiring a limited needle depth. Over time, however, the use of other non-needle catheters decreased as investigators gained familiarity with lesion titration with the SERF system.

The patients in this study were limited to those who had already failed a prior ablation. It is unclear whether the efficacy or safety would be different in patients receiving a first-time SERF ablation. All of the patients also had ischemic or non-ischemic cardiomyopathy. It is unknown if this approach would be equally useful in patients with focal intramural arrhythmias in the absence of structural heart disease, or other VT etiologies and less severe ventricular dysfunction. Further study is required to define the optimal means to select endocardial deployment sites for ablation of intramural VT substrate. Finally, the needle electrode studies were performed by highly skilled investigators. It therefore remains unclear, if those clinicians undertaking the original pre-needle ablations, before the ablations described here, would have had equivalent outcomes as seen by the experienced investigators in this needle ablation study, were it performed first.

We targeted VTs designated as clinically relevant by the investigator as the preferred approach. Limitations in defining clinical VTs are well recognized in patients with ICDs that generally terminate the arrhythmia before an ECG can be obtained². Finally, although the ablation of all inducible VTs in this population might be considered desirable, and might be associated with a lower risk of recurrence than in other studies⁴⁰, it is difficult to predict longer-term outcome. The extent to which recurrences are consistent with previously identified clinical VTs versus new or previously identified non-clinical VTs is difficult to assess, partially given the limitations of VT characterization from ICD electrograms. Long-term studies with better 12-lead detectors will be required to clarify the utility of this surrogate endpoint for completely predicting late outcome. Non-inducibility is not necessarily the only reliable predictor of long-term outcome, particularly given the inability to parse out what VT is present upon recurrence based solely on device interrogation.

Conclusion

This study demonstrates effective performance of a novel catheter ablation method in a population refractory to previous, conventional catheter ablation. The targeted inducible VT was eliminated acutely in nearly all patients and during follow-up VT was eliminated in more than 50% of patients with substantial reduction in VT burden in the remaining patients. These initial efficacy data are beyond that anticipated from a further standard RF ablation. This method may obviate the need for epicardial ablation and accompanying risk. To clarify this, larger studies will nevertheless be required.

What is Known:

• Ventricular tachycardia is common in patients with coronary artery and idiopathic myocardium disease

• The ability to successfully eliminate arrhythmia sites is difficult depending on their location

• Sometimes this is because the actual arrhythmia site is internal at the middle or outside of the heart muscle

What the Study Adds:

• The needle electrode can be inserted 6–8mm into tissue

• Arrhythmia sites in the mid-myocardium and epicardium are readily ablated

• 5–20mm large lesions are typically created

• Excessive ablation with this system can create heart muscle injury

Nonstandard Abbreviations and Acronyms

AAD

Anti-arrhythmic drugs

AE

Adverse events

ATP

Anti-tachycardia pacing

BPM

Beats per minute

CTA

Computed tomography (angiographic delayed enhancement)

ECG

Electrocardiogram

EF

Ejection fraction

ICD

Implantable cardioverter defibrillator

LV

Left ventricular

MI

Myocardial infarction

MRS

Modified Rankin Scale

NICM

Non-ischemic cardiomyopathy

NYHA

New York Heart Association

RF

Radiofrequency

RFA

Radiofrequency catheter ablation

SA

Sinoatrial

SAE

Serious adverse event

SHD

Structural hear disease

STEMI

ST elevation myocardial infarction

VF

Ventricular fibrillation

VT

Ventricular tachycardia