Questioning the preference for dual- vs. single-chamber implantable defibrillator in primary prevention implantable cardioverter-defibrillator recipients

This editorial refers to ‘Dual- vs. single-chamber defibrillators for primary prevention of sudden cardiac death: long-term follow-up of the Défibrillateur Automatique Implantable—Prévention Primaire registry’ by P. Defaye et al., on pages 1478–1484. Everything should be made as simple as possible, but not simpler. The clinical value of the implantable cardioverter-defibrillator (ICD) for the primary prevention of sudden cardiac death has been proven in selected patients with ischaemic or non-ischaemic cardiomyopathy.1,2 However, the decision whether to use a single- or a dual-chamber device has not been addressed in the randomized trials evaluating ICD efficacy. The majority of patients enrolled in these randomized trials evaluating ICD efficacy received single-chamber devices. The addition of an atrial lead in dual-chamber ICDs may offer theoretical benefits over single-chamber devices, including enhanced arrhythmia discrimination algorithms, reduction in inappropriate shocks, and the potential to reduce the rate of hospitalization and mortality. On the contrary, the placement of an atrial lead may involve a longer procedure time and higher risk of complications. It is in this context that the study by Defaye et al. in this issue of EP-Europace is of particular interest.3 The authors analysed data of the DAI-PP registry, which consecutively enrolled 1258 single-chamber and 1250 dual-chamber ICD recipients at 12 French medical centres between 2002 and 2012. They demonstrated that perioperative complications were more frequent (12.1 vs. 8.7%, P = 0.007) with dual-chamber devices, without significant differences in appropriate, inappropriate ICD therapy, or all-cause mortality. The higher rate of perioperative complications with dual-chamber devices corroborates with previous findings from the National Cardiovascular Data Registry (NCDR) ICD of the U.S. and a Canadian retrospective cohort study.4,5 The findings of this study have important clinical implications. In current clinical practice, dual-chamber devices are often placed for enhanced arrhythmia discrimination between supraventricular and ventricular tachyarrhythmias to prevent inappropriate shocks. However, a meta-analysis demonstrated that the proportion of patients experiencing inappropriate ICD therapy was not different between single- and dual-chamber devices.6 In addition, with recent developments of novel ICD programming, concerns over a high rate of inappropriate ICD shock in single-chamber ICD patients have to be revisited. The Multicenter Automatic Defibrillator Implantation Trial: Reduce Inappropriate Therapy (MADIT-RIT) indicated that novel programming with either a high-rate cut-off (2.5 s delay before the initiation of therapy at a heart rate of ≥200 bpm) or a delayed detection for arrhythmias (60 s delay at 170–199 bpm, a 12 s delay at 200–249 bpm, and a 2.5 s delay at ≥250 bpm) is associated with a significant decrease in inappropriate ICD therapy.7 The 1-year cumulative probability of inappropriate ICD therapy was 4% in the high-rate arm and 5% in the delayed therapy arm,7 with inappropriate ICD shock rates as low as 3% in both arms (non-published data). In MADIT-RIT, only dual-chamber ICD devices were implanted; however, it is conceivable that the potential benefit of enhanced arrhythmia discrimination algorithms to reduce inappropriate therapy in dual-chamber ICDs becomes marginal for arrhythmias over 200 bpm, especially with prolonged detection time. Data from MADIT-II suggest that inappropriate ICD therapies culminate in heart rate zones below 200 bpm. Although MADIT-RIT did not include patients with single-chamber ICD devices, a recent sub-study from Multicenter Automatic Defibrillator Implantation Trial: Cardiac Resynchronization Therapy (MADIT-CRT) may shed further lights on the very question Defaye et al.3 investigated in the current issue of Europace. Ruwald et al. reported data on rates of inappropriate ICD therapy, complications, and hospitalizations in 294 single-chamber ICD patients and 410 dual-chamber ICD patients.8 In this cohort, still using conservative ICD programming, there was no significant difference in inappropriate ICD therapy, and there was a trend towards an increased complications rate in the dual-chamber ICD cohort (9 vs. 1 lead dislodgement in the dual- vs. single-chamber ICD groups). The current study by Defaye et al.3 further corroborates these findings showing not only a lack of benefit of dual-chamber ICDs in terms of inappropriate ICD therapy reduction but also a significant increase in complication rates, mainly due to lead dislodgement and pocket haematomas, when compared with the single-chamber ICD patients. Imposing additional morbidity with an increased complication rate, however, did not increase all-cause mortality, or cardiovascular mortality in this subgroup. Interestingly, there was a higher incidence of battery depletion in the dual-chamber ICD group compared with the single-chamber ICD patients, indicating potentially higher risk of infections and pocket haematomas in the long-term, not even mentioning related costs.9 The findings by Defaye et al.3 are clinically important, but they must be viewed in the context of important caveats.2 Retrospective registries have limited abilities to fully capture important clinical covariates (e.g. Class I and IIa indications for cardiac pacing, device programming, procedure time) and potential confounders (e.g. right ventricular pacing percentage in the present study) and may present with potential selection bias.10 The investigators did not provide information on the number of dual-chamber patients who were aborted cardiac resynchronization defibrillator (CRT-D) recipients in whom left ventricular leads could not be placed. This is very likely as QRS duration >150 ms (20 vs. 8%, P < 0.0001) and NYHA ≥III (26 vs. 21%) were more prevalent among dual-chamber ICD patients. In contrast, real-life registries are essential to extend general applicability of findings of randomized clinical trials. In summary, the data presented by Defaye et al.3 further strengthen prior observations showing higher perioperative complication rates and generator replacements in dual-chamber ICD recipients when compared with single-chamber ICD patients. Dual-chamber devices do not appear to offer clinical benefit over single-chamber devices with regard to all-cause mortality and prevention of inappropriate ICD shocks. Such findings warrant careful assessment of potential risks/benefits of dual-chamber ICD implantations in at-risk cardiac patients without Class I or IIa pacing indications, to select the best ICD treatment (single- vs. dual-chamber ICD) to improve survival without an increase in perioperative complications or frequent device replacements. Conflictof interest: none declared.