https://orcid.org/0000-0003-0615-9525
Subcutaneous implantable cardioverter-defibrillator (S-ICD).
This editorial refers to ‘Subcutaneous implantable cardioverter-defibrillators: long-term results of the EFFORTLESS study’, by P.D. Lambiase et al., https://doi.org/10.1093/eurheartj/ehab921.
Beyond complexity lays simplicity
A. Einstein
The advent of the subcutaneous implantable cardioverter defibrillator (S-ICD) in clinical practice was meant to provide the well-established traditional transvenous implantable cardioverter defibrillator (TV-ICD) with potentially advantageous solutions. Among them were an easier implantation technique, a simpler algorithm for life-threatening event detection only, and no need for placement of ICD components in the heart and circulation. These solutions would, in the inventors’ view, simplify the task of delivering life-saving therapy while removing some redundant technology in TV-ICDs.
Awareness of unfavourable implications of redundant TV-ICD technology was consistently raised after publication of MADIT-RIT,1 a study showing that simplification of programming and intervention reduced inappropriate therapy and mortality. The results of MADIT-RIT opened the way to new standards of TV-ICD treatment2 and created a favourable context in which a closer alignment between TV-ICD and S-ICD programming could be investigated.
The S-ICD implantation technique consists of subcutaneous or submuscular placement of the generator in the left midaxillary line at the level of the fifth and sixth intercostal spaces. and subcutaneous placement of the tunnelled lead from the generator pocket to the left of the sternum. Because the S-ICD requires accurate discrimination between the QRS complex and the T-wave (sensing analysis), pre-implant ECG screening is recommended to prevent implantation in patients at risk of oversensing. In addition, due to the lack of pacing capability, the S-ICD is precluded in patients requiring bradycardia, tachycardia, or biventricular pacing at the time of implantation.
The S-ICD underwent 10 years of pre-clinical investigation3–4 followed by the Investigational Device Exemption (IDE) trial.5 This was a prospective multicentre trial that adopted objective performance criteria conducted in cooperation with the US Food and Drug Administration (FDA). The high efficacy (100%) and safety rates (92.1%) in 314 study patients well exceeded the pre-determined performance goals, and ultimately led to FDA approval of S-ICD commercialization.
Initial clinical experience with S-ICDs was conducted in cohorts that were younger, with fewer comorbidities and more preserved left ventricular function than those commonly eligible for TV-ICD therapy. Several factors probably contributed to the adoption of such conservative indications, including unsuitability based on pre-implant QRS-T wave sensing analysis, medical concern that patients will later need bradycardia, tachycardia, or resynchronization pacing, and delayed third-party insurance. Early contributions proved that the high standards of acute conversion could be replicated in the clinical setting (97.7% first shock conversion efficacy at 18-month follow-up).6 The high inappropriate shock (19.0%) and complication rates (17.0%) observed in the first patients were mitigated rather early and quite consistently (7.0% and 10.0%) in the subsequent patients with the help of a software upgrade, change of sensing vectors, and operator experience.6
Following commercial release of the S-ICD, post-approval registries were designed to document clinical, system, and patient outcomes. They included the S-ICD System Post-Approval Study (PAS),7 conducted in US patients between 2013 and 2016, and the Evaluation oF FactORs impacTing cLinical outcome and cost EffectiveneSS of the S-ICD (EFFORTLESS) registry,8 conducted in patients from outside the USA between 2011 and 2016. By providing accurate observation in >2500 patients, these studies have largely contributed to refine our knowledge and guide subsequent development of the S-ICD system.
In this issue of the European Heart Journal, Lambiase et al. present the long-term outcome results of the EFFORTLESS S-ICD registry.9 The study describes the 5-year outcome of 994 patients from 11 countries receiving early-generation S-ICDs and expands the observations made at 1-year and 3-year follow-up in two prior publications.10,11 In EFFORTLESS, patients were enrolled prospectively and retrospectively. Specific contraindications included need for bradycardia, tachycardia, or resynchronization pacing at time of implantation. S-ICD programming was set at the investigator’s discretion. Patient demographics reflected the conservative standards of early S-ICD populations. Age at implantation was 48.4 ± 17 years and ejection fraction 0.43 ± 0.18, with fewer than half of the patients presenting with ischaemic or dilated cardiomyopathy, ∼20% with channelopathies, and <10% with New York Heart Association (NYHA) class III/IV.
The study results confirm the high S-ICD standards of efficacy reported in previous publications.10,11 Of 984 implanted patients, appropriate shock therapies were delivered in 146 (14.8%), with 98.1% of 310 episodes being successfully converted. Survival free from an appropriate shock decreased linearly with time, although slopes significantly diverged between patients who had a shock already during the first year (at higher risk of subsequent shock) and those who did not. Regression analyses showed that neither follow-up duration nor rhythm type significantly affected shock efficacy. Efficiency of therapy delivery was also not affected, as time to therapy at 5 years (17.9 s) did not differ from that recorded at 1-year follow-up (17.1 s). These results are reassuring as they prove integrity of the sensing, charging, and therapy delivery aggregate over long-term follow-up in first-generation S-ICDs.
The 5-year survival free of complications was 84.8%. Most complications (5.6%) occurred during the first 60 days after implantation. Thereafter, the curve showed a lower degree of decline as time progressed, with an incidence of complications of 8.9% at 1-year and of 15.2% at 5-year follow-up. The most common complications requiring intervention were infection (3.2%), element movement or suboptimal positioning (2.5%), inappropriate shock (2.4%), and erosion (2.3%). Other complications included patient discomfort (1.1%), haematoma (0.9%), and other technical complications in the remaining cases.
Inappropriate shocks occurred in 155 patients (15.7%), totalling 328 events. Of them, 69 (7.0%) experienced just one episode, and 81 (8.2%) at least two episodes. Cardiac oversensing accounted for the most common cause (68.3%), followed by false rhythm detection (18.7%), and non-cardiac oversensing (16.8%). Changes in programming did not affect the risk of recurrent inappropriate shock. It is noteworthy that inappropriate shocks in EFFORTLESS were experienced in the absence of SMART Pass detection technology.
Early battery depletion was observed in 0.8% of patients undergoing device replacement at 4.8 years from implantation. No other complications were related to advisories affecting the model lead or low-voltage capacitor.
Other relevant findings in EFFORTLESS include a very low proportion (2.0%) of patients requiring subsequent pacing, compatibility with previously implanted pacemakers in 30 patients, a risk of infection or erosion in 139 patients with previously infected TV-ICDs (6.6%) similar to the risk in all remaining patients (5.5%), and absence of endocarditis, chronic lead malfunction, and death related to device malfunction (in 0.7% of patients, though, the cause of death was unknown).
No prospective registries have been available until recently to document very long-term outcomes in recipients of TV-ICDs. In MADIT-RIT1 and ADVANCE III,12 two prospective trials investigating the benefit of simplified programming and intervention, 2 and 5% of recipients experienced inappropriate shocks at 1.4- and 1.0-year follow-up. These figures represent reliable standards for modern TV-ICDs, albeit with no long-term data available. One-year infection rates were 1.4, 1.5, and 2.0% for single, dual, and biventricular TV-ICDs in a large cohort of Medicare patients.13 The 1-year infection rate at 2.5% in EFFORTLESS10 favourably compares with TV-ICD figures, whereas the incidence of inappropriate shock caused by cardiac oversensing appears well above TV-ICD standards. Overall, data from EFFORTLESS need to be considered in light of the healthier population investigated, as compared with usual TV-ICD populations and the exclusion of patients requiring pacing therapy. The potential benefit of antitachycardia pacing with fast ventricular tachycardias,14 possibly resulting in reduced shock rates by TV-ICDs, needs to be evaluated in comparison with the deleterious effect that repetitive or delayed interventions may have on inappropriate therapy and mortality
In recent years, new-generation S-ICDs have been tested. UNTOUCHED enrolled patients with ejection fraction ≤0.35 using standardized (dual zone) programming and improved sensing algorithms.2 In a 1111 prospective patient cohort with characteristics more aligned to TV-ICD populations, conversion and complication-free rates were 98.5% and 92.7% at 18-month follow-up. Inappropriate shocks at 1 year were 4.1% with second- and 2.4% with third-generation devices. In PRAETORIAN,15 the S-ICD was non-inferior to the TV-ICDs with respect to device-related complications, inappropriate shocks, and mortality. Enrolment in this study took place between 2011 and 2017, a time period during which the contribution by new-generation S-ICD technology was not available or insufficiently mastered.
The study by Lambiase et al.9 shows that early-generation S-ICDs were strong from the start, and successfully passed the powerful test of time. Also thanks to the contribution by second- and third-generation technologies, the S-ICD system continues to evolve with appreciable results. At present, automatic QRS-T sensing analysis enables suitability for S-ICD in >90% of screened patients. Rapid, ‘smart’ evolution of algorithm technology has reduced inappropriate shock rates to <2.5% at intermediate follow-up. Magnetic resonance imaging- (MRI) compatible technology, remote monitoring, and algorithms for detection of atrial fibrillation have meanwhile been integrated in currently commercialized units. Results from recent studies encourage the spread of indications to previously neglected patient categories. The risk of a later need for bradycardia, tachycardia, and resynchronization pacing in patients with no such indication at the time of implantation now appears negligible and encourages selection of S-ICD vs. TV-ICDs in larger populations. Future studies are necessary to consolidate current evidence, test upcoming improvements of S-ICDs, and assess the cost-effectiveness of S-ICDs compared with TV-IVDs. Results from these studies will help to refine the coming field of action of S-ICDs in the everlasting battle of man against sudden cardiac death.
Conflict of interest: R.C.: Co-founder of Cameron-Health, has owned equity and intellectual property rights with Cameron-Health.
References
Author notes
The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.
