Subcutaneous implantable cardioverter defibrillators

INTRODUCTION — Sudden cardiac death (SCD) resulting from cardiac arrhythmia is the world's leading cause of cardiovascular mortality, accounting for over 50 percent of cardiovascular deaths worldwide. Implantable cardioverter-defibrillators (ICDs) have been shown in numerous large clinical trials to reduce mortality from SCD in selected populations.

Traditionally, ICD systems consist of a pulse generator, typically placed in the pectoral region, and one or more leads that extend from the pulse generator to the myocardium via a transvenous route. However, conventional transvenous ICD (TV-ICD) systems have drawbacks, which include risks associated with insertion (eg, cardiac perforation, pericardial effusion) and persistence in the intravascular space (eg, endocarditis, lead fracture).

The subcutaneous ICD (S-ICD) has been developed in an attempt to eliminate some of the limitations of TV-ICD systems by avoiding endovascular access entirely (figure 1 and image 1).

The S-ICD system, including its indications, efficacy, complications, and our approach to choosing the proper candidates for the S-ICD, will be discussed in detail here.

TV-ICD systems, including their potential complications, are discussed separately. (See "Implantable cardioverter-defibrillators: Overview of indications, components, and functions".)

S-ICD COMPONENTS AND CAPABILITIES — Similar to a standard transvenous ICD (TV-ICD), the S-ICD is composed of a pulse generator and a lead (figure 1 and image 1) [1]. In contrast to a standard TV-ICD, the pulse generator and lead have different characteristics and are placed in different positions. Both the pulse generator and the lead are larger than traditional TV-ICD systems. The pulse generator (figure 1) is implanted in a subcutaneous (or deeper) pocket in the lower left lateral or posterolateral thoracic position, and the lead is tunneled in the subcutaneous tissue from the pulse generator to a position along the left parasternal margin (picture 1). These characteristics of the S-ICD lead to differences in how it senses and treats ventricular tachycardia.

●Rhythm detection – The cardiac rhythm is detected via a wide bipole between the two sensing electrodes or between one of the sensing electrodes and the pulse generator [2]. The electrograms generated from these vectors are similar to the those produced by a surface ECG.

●Delivery of VT shocks – S-ICDs can deliver a maximum shock of 80 joules; however, in conversion testing, a successful conversion with 65 joules is considered to provide an adequate safety margin. Because of the higher energy requirement for defibrillation, and the relatively smaller experience with S-ICD compared with TV-ICD, conversion testing is generally recommended with all newly implanted S-ICDs. Despite this recommendation, conversion testing rates remain relatively low. In a study of 7900 patients in the National Cardiovascular Data Registry with S-ICD implanted between September, 2012 and April, 2016, only 71 percent underwent conversion testing [3]. In general, the need for defibrillation threshold (DFT) testing for TV-ICD is much less imperative. (See "Implantable cardioverter-defibrillators: Overview of indications, components, and functions", section on 'Defibrillation threshold testing'.)

The device delivers an 80-Joule shock for defibrillation of ventricular tachyarrhythmias including monomorphic ventricular tachycardia (VT), polymorphic VT, and ventricular fibrillation (VF). If VT or VF persists following the initial shock, the device will reverse polarity between the electrodes and deliver subsequent shocks. The S-ICD will deliver a maximum of five shocks for a single episode of a ventricular arrhythmia. If more than 3.5 seconds of asystole occurs following a shock, the S-ICD can deliver 30 seconds of demand pacing at a rate of 50 beats per minute. During an event, the S-ICD will store the electrocardiogram (ECG) tracing for subsequent review [2].

●Sensing – The sensing algorithm for S-ICD varies substantially from the algorithms used by TV-ICDs [4]. In general, arrhythmia detection and capacitor charging (to a much higher output) for S-ICD take longer than for TV-ICD.

The S-ICD can be implanted without the use of fluoroscopy by using anatomic landmarks to guide proper positioning. In practice, brief fluoroscopy is often used to guide placement of the lead as well as the generator. The mean procedure time for implantation of an S-ICD among first-time operators is 67±33 minutes in one study, and 55±23 minutes among operators who have inserted at least three S-ICDs [2]. This is comparable to the procedure length for TV-ICD at a similar time in their development, but may be longer than contemporary TV-ICD implant times for experienced operators [5]. The results from a small, nonrandomized observational series of patients suggest that DFT testing may not be required for all patients receiving the S-ICD, although additional data are required to reproduce and validate this result [6]. (See "Implantable cardioverter-defibrillators: Overview of indications, components, and functions", section on 'Defibrillation threshold testing'.)

Initially, the S-ICD generator was placed subcutaneously in the mid to posterior axillary region. A posterior generator placement improves the shocking vector. Many implanters have moved to an intermuscular implant (between the serratus anterior and latissimus dorsi muscles) or submuscular implant. Advantages of these deeper implants include a lower shock impedance, higher likelihood of defibrillation success, and, in lean patients, a better cosmetic outcome. The implant procedure has evolved from three incisions (large pocket incision and smaller xiphoid and upper sternal incisions) to two, with the upper sternal incision usually omitted [7,8].

INDICATIONS, CONTRAINDICATIONS, AND AN APPROACH TO SELECTING THE S-ICD — S-ICDs were designed to address the limitations of conventional transvenous ICD (TV-ICD) systems, such as the need for vascular access. However, because of its unique capabilities and associated limitations, the S-ICD is not the best option for all patients requiring an ICD.

The number of patients who might potentially benefit from the S-ICD rather than a standard TV-ICD is not known (picture 2). In a retrospective single-center cohort study of 1345 patients who underwent ICD implantation for both primary and secondary indications, 463 patients (34 percent) received antitachycardia pacing as a therapy or developed an indication for bradycardia pacing or cardiac resynchronization therapy which cannot be performed using the S-ICD [9]. However, at five years of follow-up, 55 percent of the cohort would have been eligible for the S-ICD. On the other hand, when patients are properly selected, very few require revision of their device to allow for anti-tachycardia pacing or bradycardia pacing.

When to consider the S-ICD — Our approach to selecting an S-ICD for a particular patient is in general agreement with recommendations from the 2017 American Heart Association/American College of Cardiology/Heart Rhythm Society guideline for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death [10], and the 2015 European Society of Cardiology guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death [11]. S-ICDs are generally considered in the following situations:

●Younger patients (eg, age less than 45 years) with anticipated need for ICD therapy spanning decades (likely requiring multiple ICD systems over time). As examples, an S-ICD system might be an appropriate consideration in patients with hypertrophic cardiomyopathy, congenital cardiomyopathies, or inherited channelopathies. (See "Hypertrophic cardiomyopathy: Management of ventricular arrhythmias and sudden cardiac death risk", section on 'Recommendations for ICD therapy'.)

●Candidates for an ICD without a current or anticipated need for bradycardia or antitachycardia pacing [12].

●Patients at high risk for bacteremia, such as patients on hemodialysis or with chronic indwelling endovascular catheters.

●Patients with challenging vascular access or prior complications with TV-ICDs [13,14].

The S-ICD may also be an appealing option in children and teenagers who require an ICD, though data in this population are limited [15,16]. Although the contemporary S-ICD generator is substantially larger than a TV-ICD generator, the implant site and technique generally mitigate the importance of the size differential.

When to avoid the S-ICD — Aside from transient post-shock demand pacing, S-ICDs provide neither antitachycardia pacing as a therapy for ventricular arrhythmias nor continuous bradycardia pacing in the event of symptomatic bradyarrhythmias. Because of this, S-ICDs should not be implanted in patients who have monomorphic VT that is known or anticipated to be responsive to antitachycardia pacing or patients with the need for bradycardia pacing. Also related to the inability to provide chronic pacemaker activity, S-ICDs are also not indicated in patients requiring biventricular pacing for cardiac resynchronization therapy [13,14].

Although there are limited published data, a small series suggests that the S-ICD can be expected to function normally within the presence of a preexisting permanent pacemaker functioning in a bipolar pacing mode [12,17]. Unipolar pacing from a coexisting device is contraindicated. Testing for device interactions is recommended at the S-ICD implant procedure.

A preimplantation surface ECG manual screening tool has also been developed to minimize the number of patients at risk for inappropriate shock (IAS) due to T-wave oversensing errors [1,18,19]. The tool identifies patients who have large and or late T-waves relative to the QRS using three vectors that mimic the device sensing vectors. ECG screening is available via automated software embedded on device programmers. Studies suggest that between 8 and 15 percent of patients are ineligible for an S-ICD due to susceptibility to T-wave oversensing and thus high risk of inappropriate shocks [18,19].

Approach to S-ICD device selection — While there are no absolute guidelines for the selection of a S-ICD over a TV-ICD system, our experts generally considered several clinical factors in deciding between the S-ICD and TV-ICD (algorithm 1):

●Does the patient have an indication for antitachycardia pacing or known to respond to antitachycardia pacing?

●Does the patient have an indication for standard transvenous pacing?

●Is the patient a candidate for biventricular pacing and cardiac resynchronization therapy?

●Is the patient relatively young with anticipated prolonged ICD therapy or multiple ICD systems in the course of one’s lifetime?

●Does the patient have other indwelling venous catheters or leads?

●Is the patient at high risk for systemic infection?

While there are no strict guidelines on the utilization of the S-ICD in place of a TV-ICD, the answers to the above questions can provide guidance to the clinician when discussing the situation with the patient. At present, battery longevity for TV-ICD substantially exceeds that of S-ICD, which presents both clinical and systems cost considerations.

S-ICD EFFICACY — One randomized trial, and several nonrandomized studies, have evaluated the feasibility of an entirely S-ICD system, which was approved for use in the United States by the FDA in 2012 [2,20-25]. Successful detection of ventricular arrhythmias ranges from 98 to 100 percent, and conversion of induced arrhythmias during defibrillation threshold (DFT) testing ranges from 95 to 100 percent (table 1), with a mean time to therapy as low as 14 seconds (slightly longer than typically seen with TV-ICDs) [1,2,20-23,26-32]. In a 2017 systematic review of 5380 patients from 16 studies, the pooled rate of successfully terminating ventricular arrhythmias was 96 percent [31].

Nonrandomized studies have generally shown efficacy and safety of the S-ICD.

●In the IDE trial, a prospective, nonrandomized, multicenter trial of patients with a standard indication for an ICD but no pacing requirement (mean age 52 years, 74 percent male, 79 percent primary prevention), 321 patients underwent S-ICD implantation and were followed for an average of 11 months [26]. Both primary endpoints were achieved, with a primary safety endpoint (180-day device and procedure-related complication free rate) of 92 percent and a primary efficacy endpoint (conversion of induced ventricular fibrillation at the time of implantation) of 100 percent among patients who completed the DFT testing protocol. Even when a sensitivity analysis was performed that assumed that all 17 of the 321 patients who did not complete DFT testing at the time of implantation would have failed defibrillation testing, the predetermined primary efficacy endpoint was still met. Following implantation, 21 patients (7 percent) received a total of 38 appropriate ICD shocks, while 41 patients (13 percent) received at least one inappropriate ICD shock.

●Data from the EFFORTLESS S-ICD Registry, an observational study of 985 patients worldwide who have received the S-ICD (average follow-up 3.1 years), have shown complication-free rates of 96 and 92 percent at 30 and 360 days, respectively, with only 8 and 12 percent of patients having received an inappropriate shock at 1 year and 3.1 years, respectively [32].

●Similar data have been reported from the S-ICD Post-Approval Study, a prospective registry involving 86 centers in the United States. Among 1637 patients who received the S-ICD, 1394 patients (99 percent) had successful termination of induced VT at the time of device insertion, with a 30-day complication-free rate of 96 percent [33].

●In the START (Subcutaneous versus Transvenous Arrhythmia Recognition Testing) trial, which compared simulated sensing performance of the S-ICD with that of standard transvenous ICDs (TV-ICDs) in 64 patients, both S-ICD and TV-ICD devices were successful in detecting 100 percent of ventricular arrhythmias [34]. In this trial, the S-ICD also had greater success in discriminating supraventricular tachycardias from VTs (98 percent S-ICD versus 76.7 percent for single-chamber TV-ICD versus 68 percent for dual-chamber TV-ICD).

Efficacy of S-ICD defibrillation can be maximized by optimal position of the device at the time of implantation [35,36]. A risk score (PRAETORIAN score) has been developed and validated based on the following determinants as identified on post-insertion posterior-anterior and lateral chest radiographs: subcoil fat, subgenerator fat, and anterior positioning of the S-ICD generator [35]. Higher amounts of fat between the coil and the sternum, higher amounts of fat between the generator and the rib cage, and more anterior generator position are associated with higher risk of failure to successfully defibrillate. A two incision implant technique with intermuscular placement (between anterior surface of serratus anterior and the posterior surface of latissimus dorsi) of the S-ICD generator has been generally adopted to achieve optimal posterior placement of the device and ideal cosmesis and patient comfort [7].

COMPLICATIONS — The S-ICD system obviates many of the mechanical complications associated with transvenous lead implantation (eg, cardiac perforation leading to pericardial effusion and cardiac tamponade, hemothorax, pneumothorax, endovascular lead infection, venous occlusion). Additionally, the solid core design of the S-ICD lead and its lack of exposure to the repeated mechanical stresses of myocardial contraction may serve to improve lead durability when compared with TV-ICD leads [4]. However, the S-ICD system does have its own potential complications, including inappropriate shocks, pocket infection, and lead dislodgement or migration [37].

Among various publications on the S-ICD, the complication rate requiring reintervention has ranged from 1.3 to 19 percent [1,2,4,20-23,25-29,31,38].

Inappropriate shocks — Inappropriate shocks (IAS) remain as one of the most common and concerning complications seen with S-ICDs, with most studies reporting an incidence ranging from 4 to 16 percent [2,21-23,26-31,39]. In a 2017 systematic review of 5380 patients from 16 studies, the pooled rate of inappropriate shocks was 4.3 percent [31]. Overall, the most common cause for IAS from S-ICDs is T-wave oversensing (TWOS), which differs from TV-ICD systems, in which most of the IAS are due to supraventricular arrhythmias or lead malfunction [27,31,40,41]. Inappropriate sensing of myopotentials from chest muscle activity may also be a source of IAS. IAS are more likely to occur in younger, physically active patients, who are also those commonly selected for placement of an S-ICD system [20,27].

The programming of an arrhythmia discrimination zone can reduce the frequency of inappropriate S-ICD shocks due to supraventricular arrhythmias [26,42]. In the S-ICD, this is called a conditional zone, in which discriminants are applied to prevent shocks due to rate only; in the faster zone, only rate is used for diagnosis. Conditional zone programming reduced the incidence of IAS caused by supraventricular arrhythmias by 70 percent (relative risk reduction) and those caused by TWOS by 56 percent [26]. In another study which compared 226 patients with dual-zone programming and 88 patients with single-zone programming, the two-year rates of freedom from IAS were 89.7 and 73.6 percent, respectively [42].

The morphology-based sensing algorithm was subsequently enhanced in an effort to reduce TWOS without sacrificing tachyarrhythmia detection. In a simulation using recorded TWOS episodes, as well as ventricular and supraventricular arrhythmias, the enhanced algorithm reduced inappropriate detection and device charging by 39.8 percent without a significant reduction in appropriate ventricular arrhythmia detections or specificity for supraventricular arrhythmias [43].

More recently, a high-pass filter (Smart Pass [SP]) was developed to further reduce IAS due to TWOS. In a study of 1984 patients, SP reduced first IAS by 50 percent and all IAS by 68 percent. There was no significant difference in rate of appropriate shocks or in time from arrhythmia onset to appropriate shock [44]. The SP feature has been available as a downloadable software upgrade to second-generation A209 devices since regulatory approval (United States: August, 2016; Europe: April, 2016), and is automatically enabled on contemporary A219 devices.

In UNTOUCHED, the IAS rate was studied in a contemporary primary prevention ICD population, utilizing contemporary S-ICD devices with standardized programming and contemporary discrimination algorithms. These 1111 patients had generation 2 or 3 devices, dual-zone programming (conditional zone 200 to 250 beats per minute), and SP enabled in 60 percent of patients. [45]. At 18 months, the incidence of IAS was 4.1 percent; this compares favorably to IAS rates in studies of contemporary TV-ICDs. Patients with generation 3 devices had a 53 percent reduction in IAS rates. The conversion success rate for appropriate discrete episodes was 98.4 percent. At 18 months, the device-related complication rate was 7.3 percent.

Pocket hematoma — The development of pocket hematoma requiring evacuation, transfusion, or extended hospital stay following S-ICD implantation is relatively low (reported rates of 1 to 5 percent) and similar to rates seen with TV-ICDs [22,23,46]. In a retrospective study of 200 patients who received the S-ICD at one of two academic medical centers, 10 patients (5 percent) developed a hematoma, with significantly greater likelihood in patients in whom antithrombotic therapy was uninterrupted or bridged with heparin (6 of 30 patients [20 percent] compared with 0 of 26 patients in whom antithrombotic therapy was stopped) [46]. Given the relatively small number of patients and events in this study, the optimal approach to management of antithrombotic and antiplatelet therapy in patients undergoing S-ICD implantation remains to be determined. However, our authors feel that the general approach to management anticoagulation with S-ICD implantation should be similar to that with TV-ICD implantation, namely avoiding the use of bridging anticoagulation with the consideration of device implantation on uninterrupted oral anticoagulation for patients at highest risk of thromboembolic events. (See "Cardiac implantable electronic devices: Periprocedural complications", section on 'Bleeding'.)

Pocket infections — While generally less concerning than infection involving TV-ICD systems, in which the indwelling venous leads pose a higher risk of systemic infection, pocket infections remain a concern with the S-ICD. Pocket infections have been noted in 1 to 10 percent of S-ICD recipients [2,21,22,26,27,31,38]. In a 2017 systematic review of 5380 patients from 16 studies, the pooled rate of pocket infection was 2.7 percent [31]. Complicated infections requiring device explantation are less frequent (1 to 4 percent of patients) [23,26,27]. Unlike the recommended course of therapy for an infected TV-ICD, S-ICD infections can be treated conservatively with a course of antibiotics and without removal of the S-ICD. Because the S-ICD device does not contain any endovascular leads, the risk of infection causing bacteremia/endocarditis is reduced, and in the event an S-ICD does require extraction, this procedure has less associated risk than transvenous lead extraction.

Lead movement — Lead dislodgement or migration had been noted to occur in 3 to 11 percent of patients in various studies [2,21,29]. Typically, lead dislodgement or migration is thought to result from vigorous physical activity occurring without adequate fixation of the parasternal lead and requires reoperation to reposition the lead [2,21]. Initially, the lead was fixed at both the xiphoid and upper sternal locations; this essentially eliminated lead dislodgement and migration [21]. The distal (upper sternal fixation) appears to be superfluous. In most patients, a single suture sleeve is used to anchor the lead near the xiphoid.

Lead failure — While it was anticipated that mechanical lead failure would be extraordinarily rare in S-ICD systems, a medical advisory was issued in December, 2020, describing a small number of lead fractures just distal to the proximal sense ring [47]. Mitigation strategies have been recommended. It was anticipated that extraction of chronically implanted S-ICD leads would be far more straightforward than for TV leads, which frequently require specialized extraction equipment due to intravascular scarring and fibrous adhesions (see "Cardiac implantable electronic device lead removal"). Though experience thus far has been limited, in some cases, fibrous encapsulation of the subcutaneous lead has made extraction more difficult than expected [48]. In a French registry describing 32 patients, simple traction was successful in 19 patients, while three required an additional incision, and nine required a mechanical sheath to break up adhesions. In one patient, lead extraction was unsuccessful; the patient had undergone subsequent coronary artery bypass graft surgery, and the lead was inadvertently entrapped in a sternal wire [49].

Other less common complications that may require reintervention may include skin erosion, premature battery depletion, or explantation due to need for antitachycardia/bradycardia pacing or a new indication for resynchronization therapy [21]. In a cohort of 55 patients with the S-ICD who were followed for a median of 5.8 years, 26 patients (47 percent) underwent device replacement, with 25 of 26 patients requiring replacement for battery depletion [50]. The median time to replacement was five years, with five patients requiring replacement due to premature battery depletion within 18 months after implantation. At five years, 71 percent of devices were still in service.

Complication rates have been shown to improve as operators and centers gain experience with S-ICD implantation. In one study of 118 patients who underwent S-ICD implantation, adverse events were more frequent in the first 15 implantations per center compared with subsequent implantations (17 percent versus 10 percent with later implantations), suggesting significantly improved outcomes with center experience [21].

COMPARISON WITH TV-ICD — One randomized trial and several nonrandomized studies have directly compared the efficacy of the S-ICD with traditional transvenous ICDs (TV-ICDs). A single nonblinded, randomized, noninferiority study compared S-ICD with TV-ICD [51]. In PRAETORIAN, 849 patients (81 percent primary prevention) who were felt appropriate for either type of device were randomized (426 S-ICD, 423 TV-ICD). The primary endpoint was a composite of device-related complications and inappropriate shocks (IAS). At median follow-up of 49.1 months, important results included:

●For the primary endpoint, S-ICD was noninferior to TV-ICD. Sixty-eight (15.1 percent) S-ICD patients and 68 (15.7 percent) TV-ICD patients had device-related complications or IAS (hazard ratio [HR] 0.99; 95% CI 0.71-1.39).

●Device-related complications occurred in 31 S-ICD patients and 44 TV-ICD patients (HR 0.69; 95% CI 0.44-1.09).

●IAS occurred in 41 S-ICD patients and 29 TV- ICD patients (HR 1.43; 95% CI 0.89-2.30). Importantly, only 22 percent of the S-ICD patients had the Smart Pass (SP) filter active at the time of their first IAS, because of the era of patient enrollment.

●The secondary endpoint of death occurred in 83 S-ICD patients and 68 TV-ICD patients (HR 1.23; 95% CI 0.89-1.70). Sudden deaths were balanced in the two groups, but S-ICD patients had more nonsudden cardiovascular deaths and many more noncardiovascular deaths than TV-ICD patients.

●The secondary endpoint of appropriate shocks occurred in 83 S-ICD patients and 57 TV-ICD patients (HR 1.52; 95% CI 1.08-2.12). This was largely due to the absence of antitachycardia pacing in the S-ICD patients; antitachycardia pacing terminated approximately half of the ventricular tachycardia episodes in the TV-ICD group.

●Similar mortality – A study from the National Cardiovascular Data Registry showed that among over 16,063 patients (mean age 73 years, 991 with S-ICD), there were no differences in mortality between patients with S-ICD and TV-ICD (incidence rates per 100 person years 12 versus 9; or in-hospital readmission 48 versus 38) [52]. Rates of complications, including device reoperation, device removal for infection, or other reasons, were also similar.

●Similar rates of infection, device failure, and inappropriate therapies – A meta-analysis of five nonrandomized studies compared outcomes in S-ICD versus TV-ICD patients in over 6400 patients. Lead complications were fewer in S-ICD patients (odds ratio [OR] 0.13; 95% CI 0.05-0.38). There were no significant differences in infections (OR 0.75; 95% CI 0.30-1.89) or system/device failures (OR 1.13; 95% CI 0.43-3.02). Inappropriate therapies were similar (OR 0.87; 95% CI 0.51-1.49), though the causes were different. S-ICD patients were susceptible to inappropriate therapies due to T-wave oversensing and noise, while in TV-ICD patients, inappropriate therapies were more commonly due to supraventricular tachyarrhythmias, including atrial fibrillation [37].

In a retrospective cohort study of 1160 patients from two hospitals who received an ICD between 2005 and 2014 (including 148 who received an S-ICD between 2009 and 2014), propensity analysis was performed on 280 patients (140 S-ICD recipients and 140 matched TV-ICD recipients) [53]. The overall complication rate was not significantly different between the two groups (14 percent for S-ICD recipients versus 18 percent for TV-ICD recipients), with the S-ICD recipients experiencing significantly fewer lead-related complications (0.8 versus 11.5 percent) but significantly greater nonlead-related complications (9.9 versus 2.2 percent). While TV-ICD patients had significantly more ICD interventions (shocks plus antitachycardia pacing; HR 2.4), there was no significant difference in the frequency of shocks (both appropriate and inappropriate) between the two groups.

FUTURE DIRECTIONS — Limitations and complications related to endocardial leads have fueled the development of both S-ICD and leadless pacing (see "Permanent cardiac pacing: Overview of devices and indications", section on 'Leadless systems'). It is likely that S-ICD will be paired with a leadless right ventricular pacemaker to allow antitachycardia pacing and bradycardia pacing. Leadless right atrial and left ventricular pacing are currently in development, raising the possibility of modular multichamber pacing and ICD systems in the future.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Heart failure in adults" and "Society guideline links: Ventricular arrhythmias" and "Society guideline links: Cardiac implantable electronic devices".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Implantable cardioverter-defibrillators (The Basics)" and "Patient education: Sudden cardiac arrest (The Basics)")

●Beyond the Basics topic (see "Patient education: Implantable cardioverter-defibrillators (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Components and capabilities of S-ICD – The subcutaneous (S-ICD) is composed of a pulse generator and single shocking coil running along the left parasternal margin (figure 1 and picture 1 and image 1). These are both implanted subcutaneously (image 2 and picture 3) without endovascular access. Implantation may be performed using anatomic landmarks without the use of fluoroscopy. (See 'S-ICD components and capabilities' above.)

●Potential drawbacks – Despite many well-documented benefits for appropriate patients, transvenous implantable cardioverter defibrillators (TV-ICDs) possess a number of drawbacks, which are most notably related to the reliance on endovascular leads. The subcutaneous ICD (S-ICD) has been developed in an attempt to minimize some of the limitations of TV-ICD systems by avoiding endovascular access entirely. (See 'Introduction' above.)

●Patient selection – Appropriate patient selection for the S-ICD is continuing to evolve. However, S-ICDs may be considered in certain subsets of patients (see 'Indications, contraindications, and an approach to selecting the S-ICD' above):

•Younger patients due to the expected longevity of the implanted leads and a desire to avoid chronic transvenous leads

•Candi dates for an ICD without a current or anticipated need for pacing (figure 1)

•Patients at high risk for bacteremia, such as patients on hemodialysis or with chronic indwelling endovascular catheters

•Patients with challenging vascular access or prior TV-ICD complications

●Indications – While there are no defined guidelines for the selection of an S-ICD over a TV-ICD system, our approach in deciding between the S-ICD and TV-ICD is based on several clinical variables (algorithm 1). (See 'Approach to S-ICD device selection' above.)

●Efficacy – S-ICDs have proven to be very efficacious, with proper arrhythmia detection rates in >99 percent of patients and successful spontaneous arrhythmia conversion rates of 88 percent on first shock (100 percent with a maximum of five shocks), both of which are comparable to rates seen with traditional TV-ICDs. (See 'S-ICD efficacy' above.)

S-ICDs do not have the capability of providing continuous pacing; therefore, S-ICDs should not be utilized for patients requiring pacing, antitachycardia pacing, or cardiac resynchronization therapy. (See 'When to avoid the S-ICD' above.)

●Complications – The S-ICD system does have its own potential complications, including inappropriate shocks, pocket infection, and lead dislodgement or migration. Inappropriate shocks appear to be the most common and concerning complication, but their frequency may be minimized by appropriate patient screening prior to implantation and appropriate device programming following implantation. (See 'When to avoid the S-ICD' above and 'Complications' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff thank Jeffrey Selan, MD, Arjun Majithia, MD, Jonathan Weinstock, MD, FACC, FHRS, and Leonard Ganz, MD, FHRS, FACC, who contributed to earlier versions of this topic review.