Author + information
- Received October 28, 2016
- Revision received December 5, 2016
- Accepted December 22, 2016
- Published online August 21, 2017.
- Joseph Y.S. Chan, MBBSa,∗ (, )
- Jacek Lelakowski, MD, PhDb,
- Francis D. Murgatroyd, MDc,
- Lucas V. Boersma, MD, PhDd,
- Jian Cao, PhDe,
- Vladimir Nikolski, PhDe,
- Griet Wouters, MScf and
- Mark C.S. Hall, MDg
- aDivision of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, Hong Kong, The Chinese University of Hong Kong, Hong Kong
- bDepartment of Electrocardiology, The John Paul II Hospital, Krakow, Poland
- cDepartment of Cardiology, King’s College Hospital, London, United Kingdom
- dCardiology Department, St. Antonius Hospital, Nieuwegein, the Netherlands
- eCardiac Rhythm and Heart Failure, Medtronic Inc., Mounds View, Minnesota
- fCardiac Rhythm and Heart Failure, Medtronic Inc., Maastricht, the Netherlands
- gDepartment of Cardiology, Liverpool Heart and Chest Hospital, Liverpool, United Kingdom
- ↵∗Address for correspondence:
Dr. Joseph Y.S. Chan, Prince of Wales Hospital, Chinese University of Hong Kong, Department of Medicine and Therapeutics, 30-32 Ngan Shing Street, Hong Kong.
Objectives This study assessed the defibrillation efficacy of the substernal-lateral electrode configuration.
Background Subcutaneous implantable cardioverter-defibrillators (ICDs) are regarded as alternatives to transvenous ICDs in certain subjects. However, substantially higher shock energy of up to 80 J may be required. Proposed is a new defibrillation method of placing the shock coil into the substernal space.
Methods This prospective, nonrandomized, feasibility study was conducted in subjects scheduled for midline sternotomy or implant of ICD. A blunted end tunneling tool was used to insert a defibrillation lead behind the sternum using a percutaneous subxiphoid approach. A skin patch electrode was placed on the left mid-axillary line at the fourth to fifth intercostal space. After ventricular fibrillation induction, a single 35-J shock was delivered between the lead and skin patch.
Results Sixteen subjects (12 males, 4 females; mean age: 61.6 ± 11.8 years) were enrolled. The mean lead placement time was 11.1 ± 6.6 min. Of the 14 subjects with successfully induced ventricular fibrillation episodes, 13 subjects (92.9%) had successful defibrillation. The 1 failure was associated with high and lateral shock coil placement. Mean ventricular fibrillation duration was 18.4 ± 5.6 s with a shock impedance of 98.1 ± 19.3 ohms. Of the 11 subjects with coil-patch electrograms, the average R-wave amplitude during sinus rhythm was 3.0 ± 1.4 mV.
Conclusions These preliminary data demonstrate that substernal defibrillation is feasible and successful defibrillation can be achieved with the shock energy available in current transvenous ICDs. This may open new alternatives to extravascular ICD therapy.
Conventional transvenous (TV) implantable cardioverter-defibrillator (ICD) systems suffer from complications related to TV leads, sometimes with serious sequelae (1). At implantation there are risks associated with TV lead insertion including pneumothorax, hemothorax and cardiac/vascular perforation (2). TV leads are not without problems in long-term observation. Premature lead failure has led to recalls and lead failure incidence has been reported to be as high as 20% at 10 years (3–5). The incidence of lead failure is particularly high among certain population, like children and young adults (6). When lead failure occurs, TV lead extraction may be necessary, a procedure associated with significant morbidity and mortality (7–10).
An entirely subcutaneous (SQ) ICD has been developed to circumvent the problems related to TV lead as well as to prevent SCD in patients with contraindications to conventional ICD systems (11–14). The SQ ICD provides a viable option for patients with venous access problems or challenging anatomy in cases of congenital heart disease. The recent European Society of Cardiology guideline recommends an SQ ICD as an alternative to a TV ICD in subjects without indications for bradycardia pacing, cardiac resynchronization, or antitachycardia pacing (15). However, SQ ICDs require substantially higher shock energy (up to 80 J may be required) and painless pacing at relatively low output is not possible. This disadvantage leads to relatively large pulse generator, limited ICD longevity (16) and inability to deliver painless backup and antitachycardia pacing. A space that can house a defibrillation lead with potentially lower defibrillation threshold and pacing options is present behind the sternum and in front of the pericardium. The aim of this study was to assess the feasibility and defibrillation efficacy of novel substernal-lateral ICD leads configuration.
This was a global, prospective, nonrandomized acute feasibility study involving 5 centers in Hong Kong, the Netherlands, Poland, and the United Kingdom. Subjects >18 years of age undergoing surgery requiring midline sternotomy or subjects undergoing TV or SQ ICD implantation were included. Exclusion criteria included anticipated high risk of stroke or infection, pacemaker dependency, previous pericarditis, prior sternotomy, hypertrophic cardiomyopathy, severe aortic stenosis, severe proximal 3-vessel coronary artery disease, or >50% stenosis in left main stem. The ethics committees and institutional review boards of each individual site locally approved the study protocol, and all subjects provided signed informed consent for trial participation.
The main endpoint of the study was the defibrillation efficacy of a 35-J shock delivered by the substernal-lateral configuration of induced ventricular fibrillation (VF). The hypothesis was that the proportion of successful shock would be >65%. The exact method was used to calculate the lower 95% confidence bound. In addition, the safety of the percutaneous introduction of a defibrillation lead into the substernal space and delivery of defibrillation shock was assessed.
Access to the substernal space
The implant tools used in the study comprised market-released components and no dedicated tools designed for defibrillation were used. A pre-gelled, self-adhesive, skin patch electrode (Physio Control Inc., Redmond, Washington) was first placed along the left midaxillary line at the fourth to fifth intercostal space. Under general or local anesthesia, a skin incision was made in the subxyphoid area and blunt dissection performed of the SQ tissue toward the sternum. A blunt stainless steel rod (Model 6996T Tunneling Tool, Medtronic, Inc., Minneapolis, Minnesota) was shaped by the investigator at the distal portion to facilitate introduction according to the subject’s anatomy. A peel-away sheath was used in conjunction with the rod and introduced into the substernal space under fluoroscopic guidance. An 8-cm superior vena cava defibrillation coil (Model 6937, Medtronic Inc.) was introduced into the sheath after removal of the stainless steel rod. The position of the defibrillation coil was adjusted under fluoroscopy (Figure 1). The skin patch and the defibrillation coil were connected to a TV ICD (Protecta, Medtronic Inc.) that was ex vivo.
VF induction and defibrillation
A diagnostic electrophysiology study catheter via femoral venous access or standard right ventricular ICD lead via subclavian venous access was placed in the right ventricle for VF induction. A single episode of VF was induced using a 50-Hz burst. After VF was confirmed by surface electrocardiogram, a 35-J shock was delivered manually by the ex vivo ICD (shock vector: patch electrode to shock coil) through a programmer (Model 2090, Medtronic Inc.). If the shock failed to terminate VF, an external transthoracic defibrillator was used for rescue. Cine images were used to analyze possible reasons of failed ICD shock. Once study testing was complete, all study components were removed and the subject’s planned surgical procedure proceeded according to standard of care. Subjects were followed through their routine post-surgery follow-up visit (5 to 50 days) to ensure that the substernal lead tunneling track used for the test shock healed appropriately.
A total of 16 subjects were enrolled between January and October 2015. Ten studies preceded midline sternotomy and 6 preceded ICD implants (Table 1). Twelve studies (75%) were performed under general anesthesia. Four subjects (25%) underwent SQ ICD implants were conducted under conscious sedation. Tunneling and lead placement was successful in all subjects. Mean lead placement time was 11.1 ± 6.6 min. Subjects with induced and treated VF were included in the primary objective. Of the 14 subjects with successfully induced VF episodes, 13 subjects (92.9%) had successful defibrillation at 35 J. The primary objective of true proportion of shock success was met with a lower bound of 70.3% (>65% with an a priori significance level of 0.05). The 1 failure was associated with a high and lateral shock coil placement with a portion of the defibrillation coil protruding outside of the superior border of the cardiac silhouette in an anteroposterior fluoroscopic view (Figure 2, right). One external rescue shock successfully converted the patient to sinus rhythm. Analysis of cine images in anteroposterior projection showed that the rest of the subjects had the coil within or only with minimal protrusion outside the cardiac border (Figure 2, left). Mean VF duration was 18.4 ± 5.6 s with shock impedance of 98.1 ± 19.3 ohms. Of the 16 subjects with VF induction testing, 1 pre-shock rhythm of monomorphic ventricular tachycardia occurred with a successful shock conversion to normal sinus rhythm (ventricular tachycardia duration 17 s with shock impedance of 125 ohms); the other subject had spontaneous termination of ventricular tachycardia just before shock. These 2 subjects were excluded from the primary objective analysis. Of the 11 subjects with coil-patch electrograms, the average R-wave amplitude during normal sinus rhythm was 3.0 ± 1.4 mV.
Three study-related adverse events were observed. One patient developed hypoxia after extubation, productive cough, and pyrexia that required antibiotic treatment. This particular patient underwent open heart surgery, which was complicated by a right lung pleura laceration from the sternotomy bone saw. This was attributed to an anatomic variation of the patient with the right pleura extending across the midline of the sternum. The second adverse event was noted in a patient scheduled for coronary artery bypass grafting who had bruising along the left internal mammary artery bed after the tunneling procedure, which prevented use of the left internal mammary artery for bypass grafting. This was the same patient who experienced unsuccessful defibrillation and had the defibrillation coil placed high and lateral. The last adverse event was an allergic reaction to the external defibrillation pad that resolved spontaneously 1 day after the procedure.
The substernal space is defined as the mediastinal area that lies between the sternum and the heart, and is composed predominantly of loose connective tissue absent of major blood vessels, thus making it relatively easy and safe to access by percutaneous methodologies. Additionally, its proximity to the heart provides an opportunity for effective defibrillation. We describe the first systematic evaluation of defibrillation efficacy in humans using a defibrillator coil within the substernal space. The study demonstrated high defibrillation efficacy with a high success rate of substernal tunneling via a percutaneous approach. In the second phase of the pivotal trial of SQ ICD, the mean defibrillation threshold determined using a step up/down protocol in 49 subjects was 36.6 ± 19.8 J (11). However, in the present study the defibrillation threshold was not determined and no direct comparison should be made between the 2 studies. Our study design is closer to the European trial arm of the SQ ICD pivotal study, which studied the implantation in 53 subjects with implant criteria of successful defibrillation of 2 consecutive induced VF episodes with 65-J shock. Out of the 53 patients, 2 needed to reverse polarity to meet the implant criteria and 1 patient did not satisfy implant criteria and resorted to a TV ICD implantation.
The adverse events related to the tunneling highlighted the importance of understanding the anatomy of the substernal space, including the relationship with the heart chambers and possible anatomic variations. The case of failed defibrillation had the tunneling tract extended outside the boundary of substernal space. This suggests the targeted location of the defibrillation lead should be within the confinement of the sternal border for effective defibrillation. The findings of this feasibility study set the stage for designing better tunneling tools and a dedicated lead design for defibrillation using a substernal-left lateral configuration.
The first-in-human experience of a defibrillation coil placed in the substernal space was performed in 3 patients with venous occlusion and 1 patient wishing to avoid implantation of a TV lead (17). All implants were successful and without complication. Safety margins between 10 and 17 J were achieved during acute implant testing. Another recent case report described a patient in whom defibrillation testing with an 80-J shock from a SQ ICD system failed. The SQ-left lateral electrode positions were converted to a substernal-left lateral configuration and induced VF was successfully detected and terminated with a 70-J shock. Recently, Sholevar et al. (18) reported on pacing success via EP catheter positioned in the substernal space. In 15 of 24 subjects with successful bipolar pacing, the median threshold was 2.9 V/10 ms. These reports support the feasibility and safety of ICD therapy with lead placement in the substernal space.
There are limitations of this study that should be considered when interpreting our findings. First, a substantial proportion of the subjects enrolled did not have severe systolic heart failure, as is typical in most primary prevention ICD recipients. The overall left ejection fraction was 47.5%. Hence, the applicability of these findings to the population with more severe heart failure will need further evaluation. Second, this study used only market-released devices for tunneling and defibrillation, which may have increased the chance of complication and failure to defibrillation. This might well be the case for the subject with defibrillation failure, which may have been avoided with a better tool for tunneling. Third, defibrillation shock was delivered manually and VF induction was performed using a TV ICD lead or diagnostic electrophysiology catheter and not through the substernal shock coil and skin patch; hence, VF detection and induction by substernal-lateral electrode configuration were not examined. Market-released TV ICD systems are designed for detection of electrograms and pacing from leads implanted TV. Thus, a totally new ICD generator design is required to serve these purposes using a substernal-lateral electrode configuration. Fourth, long-term complications relating to such configuration including possibility of mediastinitis and feasibility of percutaneous extraction were not addressed. Last, the number of subjects tested was small, but the findings of this study support further research of this novel configuration in ICD system.
These preliminary data demonstrate the feasibility and the efficiency of defibrillation using a substernal-left lateral electrode configuration. Importantly, a defibrillation lead can be placed in the substernal space safely with the use of simple tools. Findings from this study pave the way for future research aiming to improve ICD therapy.
COMPETENCY IN MEDICAL KNOWLEDGE: This systematic evaluation has shown that defibrillation via a substernal-left lateral electrode configuration is safe and efficient with a 35-J shock from a TV ICD.
TRANSLATIONAL OUTLOOK: Future studies should investigate the acute and long-term safety and performance of implant tools and defibrillation system designed specifically for substernal-lateral configuration. Such data will guide the clinical application of this novel defibrillation configuration.
The study was sponsored by Medtronic Inc., which collaborated in the execution of the trial and collected the data. Dr. Chan is a consultant for Medtronic, Inc., Boston Scientific Corp., and Biosense Webster, Inc. Dr. Murgatroyd is a consultant for St. Jude Medical, Boston Scientific Corp., and Sorin Group. Dr. Boersma is a consultant for Medtronic, Inc. Dr. Cao is an employee of Medtronic. Dr. Nikolski is an employee of Medtronic. Mrs. Wouters is an employee of Medtronic. Dr. Hall is a consultant for Medtronic, Inc., Boston Scientific Corp., and Biosense Webster, Inc. 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
- implantable cardioverter-defibrillator
- ventricular fibrillation
- Received October 28, 2016.
- Revision received December 5, 2016.
- Accepted December 22, 2016.
- 2017 American College of Cardiology Foundation
- van Rees J.B.,
- de Bie M.K.,
- Thijssen J.,
- et al.
- Hauser R.G.,
- Maisel W.H.,
- Friedman P.A.,
- et al.
- Janson C.M.,
- Patel A.R.,
- Bonney W.J.,
- et al.
- El-Chami M.F.,
- Merchant F.M.,
- Levy M.,
- et al.
- Deshmukh A.,
- Patel N.,
- Noseworthy P.A.,
- et al.
- Weiss R.,
- Knight B.P.,
- Gold M.R.,
- et al.
- Burke M.C.,
- Gold M.R.,
- Knight B.P.,
- et al.
- Priori S.G.,
- Blomström-Lundqvist C.,
- Mazzanti A.,
- et al.
- Tung S.K.,
- Bennett M.,
- Yeung-Lai-wah J.A.,
- et al.
- Sholevar D.,
- Tung S.,
- Kuriachan V.,
- et al.