**5.3 Implantable cardioverter-defibrillator**

ICD therapy has been demonstrated to improve survival in patients with heart failure as well as those with cardiomyopathy with previous cardiac arrests [35]. Therefore, most patients with VADs have already received an ICD or meet criteria for having one implanted. There are currently no randomized control trials evaluating ICD use in patients with VADs in adults or children. Studies investigating the effects of ICD therapy in patients with VAD have had mixed results. Early reports in the era of pulsatile VADs suggested an improvement in mortality rates in patients with ICDs [27, 52]. With the publication of studies evaluating adults with continuous flow VADs, three meta-analyses were published, with overlapping data, all with the finding that ICD use conferred no benefit in mortality risk [53–55]. Based on this, there is a class IIa recommendation for implantation of an ICD in patients with LVADs who have had ventricular arrhythmias in the 2017 AHA/ACC/HRS guidelines on ventricular arrhythmias [35]. There is no mention of VADs in the 2021 pediatric device consensus statement [36].

It is important to keep in mind that with the support of a VAD, ventricular arrhythmias may no longer cause hemodynamic compromise and patients may not lose consciousness, therefore a shock from a device may be felt. Adverse events in

patients with ICDs and VADs are reported in up to 30% of patients and can include changes in thresholds, inappropriate shocks caused by oversensing, and increased defibrillation thresholds [56]. Most of these patients require an ICD modification. Programming changes should be considered in the patient with a VAD to minimize shocks in the awake patient. While studies have shown significant psychological effects of being shocked by a device versus no shock in adults, this has not been replicated in pediatrics, although limited data size may have affected the ability to detect this [57, 58]. Regardless, it is in everyone's best interest to minimize pain in our patients. A single center randomized study investigated whether lengthening detection zones and increasing the use of ATP differed from nominal settings [59]. This found no difference in time to first ICD shock, but there were no harmful effects in making these adjustments. Therefore, there have been recommendations to follow this strategy with a high rate for the VF cutoff zone and the maximum number of programmable intervals [25].
