EP Research

Two-Center Experience with the Subcutaneous ICD

Abhijit Ghatak, MD, MRCP1*, Christina M. Dookhan, MD2*, Santosh K. Padala, MD3*, Rizwan Alimohammad, MD4, David A. Steckman, MD4, Henry T. Tan, MD4, James O’Brien, MD4, Mandeep S. Sidhu, MD3, Roger Carrillo, MD5, Ivan Mendoza, MD6, Assaf Tzur, MD6
1Division of Interventional Cardiology, Jackson Memorial Hospital, Miami, Florida; 2Division of Internal Medicine, Jackson Memorial Hospital, Miami, Florida; 3Division of Cardiology, Albany Medical Center, New York; 4Division of Electrophysiology, Albany Medical Center, Albany, New York; 5Division of Cardiothoracic Surgery, University of Miami Hospital, Miami, Florida; 6Department of Electrophysiology, Jackson Memorial Hospital, Miami, Florida


Abhijit Ghatak, MD, MRCP1*, Christina M. Dookhan, MD2*, Santosh K. Padala, MD3*, Rizwan Alimohammad, MD4, David A. Steckman, MD4, Henry T. Tan, MD4, James O’Brien, MD4, Mandeep S. Sidhu, MD3, Roger Carrillo, MD5, Ivan Mendoza, MD6, Assaf Tzur, MD6
1Division of Interventional Cardiology, Jackson Memorial Hospital, Miami, Florida; 2Division of Internal Medicine, Jackson Memorial Hospital, Miami, Florida; 3Division of Cardiology, Albany Medical Center, New York; 4Division of Electrophysiology, Albany Medical Center, Albany, New York; 5Division of Cardiothoracic Surgery, University of Miami Hospital, Miami, Florida; 6Department of Electrophysiology, Jackson Memorial Hospital, Miami, Florida



Background. The subcutaneous implantable cardioverter-defibrillator (S-ICD) is a novel advance in ICD technology minimizing intravascular lead complications associated with the traditional transvenous ICD. We report our initial experience with this system. Methods. We retrospectively analyzed records of all consecutive patients who had S-ICDs implanted for Class I or IIa indications for ICDs at two academic institutions between January 2014 and March 2015. Demographics, ECG characteristics, left ventricular ejection fraction (LVEF), procedural details, complications, and follow-up data were collected. Results. Thirty-six patients (42% females, mean age 46 years, mean EF 30%) had the S-ICD implanted. Induced ventricular fibrillation (n=22) was successfully terminated in all patients (100% conversion rate). The mean procedure time for patients undergoing the two-incision technique (n=7) was significantly lower than that of the three-incision technique (80.5 ± 50 mins vs 126.6 ± 32.8 mins; P=.007). The overall complication rate was 16.7%, with all complications occurring with the first 10 implants/center. During the mean follow-up period of six months, 6 successful, appropriate shock therapies were delivered in 2 patients for ventricular tachyarrhythmias and 1 inappropriate shock therapy for low amplitude signal oversensing in 1 of the 2 patients who received appropriate therapy. Conclusion. The S-ICD is an effective novel technology for sudden cardiac death prevention in patients without indications for pacing, vascular access issues, or history of bloodstream infections. The two-incision technique significantly reduces procedure time as compared to the three-incision technique. The complication rates during the early adoption period are projected to decrease with experience.

Key words: Subcutaneous implantable cardioverter-defibrillator, ventricular arrhythmias, sudden cardiac death


The implantable cardioverter-defibrillator (ICD) provides lifesaving therapy by primary and secondary prevention of sudden cardiac death to over one million patients worldwide.1 However, requirement for transvenous access and intracardiac leads are associated with short-term procedure-related complications such as pneumothorax, cardiac perforation, pericardial effusion, cardiac tamponade, and lead dislodgement.2 Long-term potential complications of classic transvenous ICD (T-ICD) systems include lead malfunction, lead fracture, lead infection with risk of endocarditis, and venous occlusion.1 The subcutaneous ICD (S-ICD) (Boston Scientific) is a novel advance in ICD technology that circumvents these potential issues by completely eliminating the transvenous and intracardiac leads.3 However, the S-ICD lacks the capacity for atrial or ventricular pacing as well as antitachycardia pacing (ATP), and as such, is not indicated in patients with these requirements. Registry data has demonstrated that S-ICDs have comparable defibrillation success rates and inappropriate shock therapy rates to T-ICD systems.4 Our report describes our initial experience with the first 36 patients who underwent S-ICD placement at two participating academic institutions: Jackson Memorial Hospital in Miami, Florida (site A) and Albany Medical Center in Albany, New York (site B).


Medical records of all consecutive patients who had the S-ICD system implanted at these two institutions between January 2014 and March 2015 were reviewed. All patients had a Class I or IIa indication for ICD implantation based on the 2012 ACCF/AHA/HRS guidelines for device-based therapy of cardiac rhythm abnormalities.5 None of the patients had an indication for pacing for symptomatic bradycardia, ATP, or cardiac resynchronization therapy (CRT). Baseline demographics including medical history, electrocardiographic (ECG) characteristics, left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) class, indication for ICD, and procedure details were collected. The reason for placement of S-ICD versus T-ICD was recorded. Length of hospital stay as well as periprocedural and post-procedural complications were reviewed. 

Prior to the implantation, patients were evaluated by an electrophysiologist, and the differences between the transvenous and S-ICDs were explained in detail. Patients were screened for suitability of the S-ICD based on body habitus and the screening template for QRS/T wave morphology as provided by the Boston Scientific protocol (Figure 1). The waveform was obtained both in supine and standing positions for 3 vector leads at various gain settings (5, 10, 20 mV). A patient was deemed suitable for the S-ICD if the ECG screening template qualified in any same lead supine and standing, at any gain, and without significant morphologic variation in the QRS complex. Either the maximal R or S wave in the QRS complex had to fit between horizontal dashed and solid lines and the width of the complex within the vertical solid lines as provided by the template box. In addition, the T wave was also required to fit within the trailing outline of the template box. Written informed consent was obtained from all patients. An institutional review board approved the study at both institutions prior to accessing the data.

S-ICD Features and Programming

The S-ICD system is comprised of a pulse generator, subcutaneous electrode, and a programmer. The shocking coil is located between the sensing electrodes, and there are three sensing vectors, with the optimal sensing vector automatically selected on the basis of the R/T ratio post implantation. The device stores this template as a part of the automated setup. The heart rate is measured as an average of four consecutively sensed intervals. Ventricular fibrillation (VF) is diagnosed when 18 of 24 events exceed the detection zone limit. All devices have dual zone programming with a conditional zone. The conditional shock zone utilizes the feature-extraction technique and is programmed at rates of 170-250 bpm to distinguish between supraventricular and ventricular tachyarrhythmias to avoid inappropriate shocks. The device has the ability to deliver up to five shocks of 80 joules with polarity reversal after the capacitors are charged with an option for post-shock bradycardia pacing for 30 seconds.


A pocket incision was created for the pulse generator at the mid axillary site between the 5th and 6th intercostal spaces. One operator routinely used left lateral fluoroscopy to identify the retrocardiac space to guide the posterior aspect of can placement. The majority of implantations (n=29, 80.6%) were done using the three-incision technique: one lateral pocket incision and two parasternal incisions. Seven implants (all at site A) were done using the two-incision technique. In this technique, the superior parasternal incision was omitted and the lead was positioned parasternally using a standard 11 or 12 French (Fr), 14 cm peel-away sheath.6 An electrode was tunneled from the pocket to the xiphoid region and from the xiphoid to the superior parasternal region. The electrode was then connected to the pulse generator, which was positioned in the pocket. Post-procedure fluoroscopy was obtained in the electrophysiology laboratory prior to closing the pocket to confirm lead and generator placement. This was later confirmed with a chest radiograph post procedure. Intraoperative defibrillation threshold testing (DFT) was performed in more than half of the patients.


All patients were seen on the day post procedure and subsequently every day until their discharge from the hospital. These patients were followed up in a device clinic within one month post implantation. Patients were followed up for six months post procedure on a routine basis. Additional clinic visits were scheduled after shock therapy or complications. Careful history taking was performed and all arrhythmic events were routinely evaluated on every clinic visit.

Statistical Analyses

Continuous variables were expressed as mean and standard deviation, while categorical variables were expressed as frequencies and percentages. Mean procedural time was calculated and compared between the two- and three-incision technique using the student’s t test. Complication rates categorized by number of implants per center were assessed using the chi-square test. A P value <.05 was considered statistically significant.


Baseline Patient Characteristics 

The baseline patient characteristics are shown in Table 1. A total of 36 patients (site A: n=15, site B: n=21) qualified for the S-ICD. The mean age was 46.6 ± 15.4 years, and there were 15 females (41.7%). The mean LVEF for the overall cohort was 30.3 ± 14.9%. Eleven patients (30.6%) had NYHA functional class I, 16 (44.4%) patients had NYHA Class II, and 9 (25%) patients had NYHA Class III symptoms. The S-ICD was implanted for primary prevention of sud-den cardiac death in 22 patients (61.1%), while 14 (38.9%) patients received it for secondary prevention. Fifty percent (11/22) of patients undergoing S-ICD implant for primary prevention had nonischemic cardiomyopathy (NICMP). Fifteen patients (41.6%) had chronic kidney disease (eGFR <60 ml/min).

Indications for selecting the S-ICD over the T-ICD are illustrated in figure 2. The leading cause for choosing the S-ICD over the T-ICD was patient preference (67%). The remainder was done secondary to technical reasons including poor venous access in 6 (16.6%) patients, prior infection from the T-ICD in 5 (13.8%) patients, and venous obstruction due to prior implanted leads in 1 (2.8%) patient. General anesthesia was used in 20 (55.6%) patients, while local anesthesia with conscious sedation was used in 16 (44.4%) patients. The mean length of the procedure was 118.7 ± 39.6 minutes. The mean procedure time for patients undergoing the two-incision technique was significantly lower than those undergoing the three-incision technique (80.5 ± 50 mins vs 126.6 ± 32.8 mins, P=.007) (Figure 3). The mean length of hospital stay for the overall cohort was 6.3 ± 8.1 days (median 1.5 days). 

DFT testing was performed in 22 of 36 patients (site A: n=4, site B: n=18) with appropriate sensing and termination of induced VF in all patients (100% conversion rate). Induced VF (21 at the time of implantation, 1 after four weeks of implantation) was terminated with a single 65-joule shock in 20 patients (91% initial success rate). Of the two remaining patients, one required a second 80-joule shock with reverse polarity for conversion to sinus rhythm. The other patient failed three shocks (65 joules, 75 joules reverse polarity, and 80 joules initial polarity after replacing the device more posteriorly) and had to be externally defibrillated. He was brought back to the electrophysiology laboratory three days later, the pocket was recreated superiorly, and a 65-joule shock with reverse polarity was successful in terminating induced VF. Four of 21 patients at site B did not receive intraoperative DFTs due to following reasons: one patient with Brugada syndrome on quinidine for VT storm could not be induced on multiple attempts; one patient had significant intraoperative bleeding from the pocket site; and two patients with atrial fibrillation had subtherapeutic INR at the time of implantation. One of these patients with atrial fibrillation was brought back after four weeks of anticoagulation and had successful termination of induced VF with a single 65-joule shock. 

Tachytherapy Device Programming 

The shock zone was set at a mean rate of 231 ± 11 bpm. All patients had an additional conditional zone programmed at a mean rate of 195 ± 8 bpm for SVT discrimination. 

Appropriate and Inappropriate Shock Therapy 

During a mean follow-up period of 182.4 ± 101.4 days, 2 (5.7%) patients received successful, appropriate shock therapies for VT/VF. Both patients received 3 shocks each for 3 discrete episodes of VT/VF. One of these 2 patients also received 1 inappropriate shock (2.8%) for low amplitude signal oversensing. 


The overall complication rate was 16.7%. There was one post-procedural death on the first implant at site A. This patient had severe three-vessel coronary artery disease (not amenable to revascularization) with significantly depressed left ventricular systolic function and an estimated EF of 10%. The patient underwent general anesthesia and had successful implantation of the S-ICD device for primary prevention. However, during reversal of anesthesia, he developed refractory hypotension and had cardiac arrest with a wide complex tachycardia. The device appropriately shocked to terminate the rhythm and the patient returned to sinus rhythm. However, despite aggressive cardiopulmonary resuscitation, the patient did not survive. The family refused an autopsy. 

A total of 3 (8.3%) patients developed post-procedural hematoma. Clinically significant hematoma developed in 1 patient that required a unit of blood transfusion post procedure and aspira-tion of hematoma 2 weeks post implant. Minor pocket hematomas were noted in 2 other patients that resolved spontaneously. One patient (2.8%) had upper sternal wound dehiscence requiring closure a month later. One patient (2.8%) had superficial wound infection treated successfully with antibiotics. There were no cases of lead or device dislodgement, premature battery depletion, device infection leading to explantation, skin erosion, or sudden cardiac death during the follow-up period. When the overall complication rates were assessed by number of implants per center, there was a significant decrease in the complication rate as the number of implants increased per center. The overall complication rate of 11.1% during the first 5 implants per center was significantly reduced to 0% after 10 implants per center (P=.027) (Figure 4). 


This report describes our initial experience with the S-ICD system in 36 patients, implanted for primary and secondary prevention of sudden cardiac death at two academic centers. The results of our experience demonstrate that S-ICD implantation is a relatively safe procedure in properly selected patients. The conversion rate of induced VF to sinus rhythm was 100% during DFT, although two patients required additional shocks or procedures to reposition the device to become successful. 
The mean length of stay in our cohort (6.3 ± 8.1 days) was slightly longer compared to that reported by a German study (5.7 ± 4.1 days).3 Reasons for prolonged hospitalization in our study included need for hypothermia protocol in the post cardiac arrest patients, coronary angiography to rule out ischemic etiology in a post cardiac arrest patient with EF <35%, diuresis in patients with decompensated heart failure and treatment of sepsis secondary to infected arteriovenous fistula and transvenous ICD leads. 

Over a mean follow-up period of six months, 100% of spontaneous VT/VF was converted to normal sinus rhythm. The rate of inappropriate shock therapy was 2.8%, which is lower than the majority of S-ICD studies published in the literature reporting up to 15% inappropriate shock therapy.7 This is likely due to the combination of rigorous use of a preoperative ECG screening tool to decrease the rate of inappropriate shocks from oversensing issues, and the use of a conditional zone for SVT discrimination. The superiority of the S-ICD for atrial arrhythmia discrimination over the T-ICD has been validated in a prospective multicenter trial (START).8 

In our cohort, we utilized both the two- and three-incision techniques. Although only 7 out of the 36 patients underwent the two-incision technique, the mean procedure time for these patients was considerably shorter than the rest of the cohort. We have noticed in our practice that the superior parasternal incision prolongs hemostasis and closing time, and hence, this technique has been accepted at site A as the standard of procedure. The two-incision technique originally described by Knops et al offers a safe and less invasive alternative for implantation of the S-ICD system.6 In addition to being aesthetically more acceptable, we believe that this technique lessens the chance of infection and patient discomfort post procedure. We had no incidence of periprocedural complication, including lead dislodgement, in these patients.

In total, 6 (16.7%) patients undergoing implant of the S-ICD system had a complication, all of which occurred during the first 10 implants per center. The overall rate of complication in our study was comparable to the rates reported in the large Dutch cohort (14%) and UK registry (17%), but slightly higher than the complication rate of 11.1% that was reported in the recent pooled analyses of two large S-ICD studies.9-11 The overall complication rate in our study was primarily driven by 3 post-procedural hematomas, of which only 1 was clinically significant requiring blood transfusion and hematoma evacuation. The rates of wound infection and wound dehiscence remained relatively low. There was 1 death (2.8%) associated with S-ICD implantation. Prior studies have reported mortality up to 2.5%.2 The patient went into irreversible cardiogenic shock during reversal of general anesthesia. Hence, we believe that patients should be selected very judiciously for conscious sedation vs general anesthesia, and general anesthesia should be initiated only for patients who are unlikely to tolerate the procedure under conscious sedation. Remarkably, there were no cases of lead or device dislodgement, device explantation due to infection or requirement of ATP, premature battery depletion, or skin erosion that were reported in prior S-ICD registries.4, 9-11 Overall, as our experience and volume increase, complication rates are projected to further decrease, as noted in our early experience with this relatively small cohort.

The most common reason for selecting the S-ICD over the T-ICD was patient preference. This pattern has also been reported in the study by Nordkamp et al9 and is likely related to multiple factors. The patients are younger and more active, which could lead to a higher chance of lead failure in subsequent years. In addition, patient-related concerns over the risk of the implantation procedure and their concern for device and lead failure, mostly fueled by the press highlights on lead recalls, influence these choices. Our study cohort differs from prior S-ICD studies in that we had 41.6% patients with chronic kidney disease (eGFR <60 ml/min) compared to none in the studies from the UK and the Dutch,9,10 and 9% in the EFFORTLESS S-ICD registry.4 The S-ICD is an attractive option for patients with chronic kidney disease, as these patients commonly have vascular access issues and are at high risk for infections.

Although there are multiple benefits, S-ICDs are not exempt from limitations. Lack of permanent pacing and ATP therapy are the two major drawbacks. The S-ICD also has a longer time to therapy after detection of arrhythmia as compared to the T-ICD because of the need to charge up to 80 joules.10 This has not been reported to influence the clinical outcome, and in fact, may reduce unnecessary shock by allowing spontaneous resolution of the tachyarrhythmia as reported by the MADIT-RIT study.11 In addition, higher cost as compared to the T-ICD, lack of remote monitoring capabilities, MRI non-compatibility, a relatively shorter battery life of 5 years, and lack of long-term lead integrity data are some of the other disadvantages of the S-ICD system.

As clinicians and implanters, we have to be cognizant of the advantages and disadvantages of the S-ICD system while counseling patients pre-implantation. We should also be aware that currently there are no randomized control trials comparing the efficacy of the T-ICD system vs the S-ICD system, attesting to the non-inferiority of the S-ICD system. Hence, pending the results of the first randomized, multicenter PRAETORIAN trial, which is designed to compare the efficacy of the S-ICD with the T-ICD,12 this device should be used very selectively in an appropriately indicated patient population.


Our study was an observational, two-center retrospective study. Our cohort size was relatively modest; however, in this modest sample size with limited operator experience, we have been able to show the safety of this device implantation. Long-term performance and safety of the device was not assessed. 


This paper describes our clinical experience with the first 36 patients who received the S-ICD implant for primary and secondary prevention of sudden cardiac death. In summary, the S-ICD is an effective novel technology for primary and secondary prevention of sudden cardiac death in selected patients. The complication rates during the early adoption period are projected to decrease as experience increases. The two-incision technique significantly reduces the procedure time as compared to the three-incision technique. Patients without indications for bradycardia pacing, ATP, CRT, and patients with vascular access issues and history of bloodstream infections, are ideal candidates for this technology. In summary, the system can be safely implanted in a real-world setting for appropriately screened and selected patients.

Disclosures: The authors have no conflicts of interest to report regarding the content herein.   


  1. Aziz S, Leon AR, El-Chami MF. The subcutaneous defibrillator: a review of the literature. J Am Coll Cardiol. 2014;63:1473-1479.
  2. Weiss R, Knight BP, Gold MR, et al. Safety and efficacy of a totally subcutaneous implanta-ble-cardioverter defibrillator. Circulation. 2013;128:944-953.
  3. Köbe J, Reinke F, Meyer C, et al. Implantation and follow-up of totally subcutaneous versus conventional implantable cardioverter-defibrillators: a multicenter case-control study. Heart Rhythm. 2013;10:29-36.
  4. Lambiase PD, Barr C, Theuns DA, et al. Worldwide experience with a totally subcutaneous ICD: early results from the EFFORTLESS S-ICD registry. Eur Heart J. 2014;35:1657-1665.
  5. Epstein AE, DiMarco JP, Ellenbogen KA, et al. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart As-sociation Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2013;61:e6-75.
  6. Knops RE, Olde Nordkamp LRA, de Groot JR, Wilde AAM. Two-incision technique for implantation of the subcutaneous implantable cardioverter-defibrillator. Heart Rhythm. 2013;10:1240-1243.
  7. Bongiorni MG, Viani S, Zucchelli G, Di Cori A, et al. Subcutaneous implantable cardiac defibrillators: indications and limitations. Curr Heart Fail Rep. 2015;12:79-86. 
  8. Gold MR, Theuns DA, Knight BP, et al. Head-to-head comparison of arrhythmia discrimina-tion performance of subcutaneous and transvenous ICD arrhythmia detection algorithms: the START study. J Cardiovasc Electrophysiol. 2012;23:359-366.
  9. Olde Nordkamp LR, Dabiri Abkenari L, Boersma LV, et al. The entirely subcutaneous implantable cardioverter-defibrillator: initial clinical experience in a large Dutch cohort. J Am Coll Cardiol. 2012;60:1933-1939.
  10. Jarman JWE, Todd DM. United Kingdom national experience of entirely subcutaneous implantable cardioverter-defibrillator technology: important lessons to learn. Europace. 2013;15:1158-1165. 
  11. Burke MC, Gold MR, Knight BP, et al. Safety and efficacy of the totally subcutaneous implantable defibrillator: 2-year results from a pooled analysis of the IDE Study and EFFORTLESS Registry. J Am Coll Cardiol. 2015;65:1605-1615. 
  12. Moss AJ, Schuger C, Beck CA, et al. MADIT-RIT Trial Investigators. Reduction in inappro-priate therapy and mortality through ICD programming. N Engl J Med. 2012;367:2275-2283. 
  13. Knops RE, Bardy GH, Blaauw Y, et al. Rationale and design of the PRAETORIAN trial: a prospective, randomized comparison of subcutaneous and transvenous implantable cardioverter-defibrillator therapy. Am Heart J. 2012;163:753-760.