**8. Electromagnetic interactions between implantable cardioverter defibrillator and left ventricular assist devices**

Both Heartware and HeartMate III (HM3) ventricular assist devices developed an electromagnetic interference (EMI) that can it make impossible to interrogate ICD with external programmers.

EMI can be explained by a pulse-width modulator (PWM), which can emit frequencies of 8 kHz, the same frequency that Abbot and Biotronik ICD/CRT programmers emit to start reading ICD. Distance between the LVAD and the ICD programmer and the speed of the LVAD rotor can also interfere [43]. HM3 manufacturing website itself provides information on possible incompatibilities (**Table 4**) [44].


#### **Table 4.**

*Adapted from Cardiovascular Abbott. The difficulty or inability to communicate with the external programmer has been reported for or may occur with the following manufacturer families by using HM3.*

#### *Implantable Cardioverter-Defibrillator Use in Patients with Left Ventricular Assist Devices DOI: http://dx.doi.org/10.5772/intechopen.109396*

Schnegg et al. [43] analyzed *in vitro* the programming of 24 explanted ICDs from several manufacturers in the presence of a running LVAD (Heartware and HM3 LVAD). Heartware LVAD interaction was only observed in the case of Biotronik and Microport devices if the distance was shorter than 6 cm, while HM3 LVAD only Medtronic ICD devices showed no interaction (needing up to 18 cm with some Biotronik devices).

Three different strategies were tested to improve connectivity between the ICD programmer and ICD: (1) Pseudo-faraday cage to achieve electromagnetic isolation of the ICD. The isolation of the ICD device with a metal pan (pseudo-faraday cage) was not effective in those devices that had previously failed (Boston and Biotronik devices); (2) LVAD parameters modification (above all speed of rotor). Once the speed of HM3 was significantly changed (−2000 rpm, +1000 rpm), communication improved up to 45% of total devices, but 55% remained unchanged; and (3) to increase the distance between LVAD and ICD by separating the arm from the thorax. This latter option, separating the arm from the thorax by bringing the hand over the head to increase the distance between ICD and LVAD, was the most effective measure. An increase of 3 cm was achieved. This posture should be maintained for the first 2–10 seconds, and then the arm can be repositioned. Alternatively, the ICD may be removed to the contralateral side as the last solution, but given the doubtful benefit of therapies in LVAD patients, it is not usually carried out [43].

A retrospective single-center study carried out by Yalcin et al. confirmed that the EMI from the Heartmate II LVAD was only present in patients with a St Jude/Abbott device (6 of the 23 St Jude/Abbott devices. In HM3 patients, EMI was mainly present in patients with Biotronik devices: four out of the 18 and only one patient with a Medtronic device. While initial interrogation of these devices was not successful, none of the 11 cases experienced pacing inhibition or inappropriate shocks [45].

Although, two studies reported a decrease in ventricular sensing with a smaller R-wave amplitude, an impedance decrease, and a capture threshold increase after LVAD placement that could lead to failure of ventricular dysrhythmias sensing, capturing failure, and inappropriate pacing [46–48]. It should be noted that the proximity of the ICD to the LVAD controller did not affect the programming values of the ICD or the shock therapies [43, 46, 49].

Black-Maier et al. describe several major findings in a systematic review of subcutaneous-ICD after LVAD implantation: ventricular sensing amplitudes reduction, EMI appears in primary and secondary vectors that lead to inappropriate shocks (above all in the postoperative period) but rarely in alternate vector, and parameters improved spontaneously during follow-up without need for device revision or extraction [50]. However, Lopez Gil et al. [51] describe a case of a 24-year-old man implanted with an S-ICD because of idiopathy dilated cardiomyopathy and self-limiting sustained VT. After the placement of an HM3, the S-ICD became useless because of inadequate sensing due to EMI and reduced QRS voltage.
