Vivek Y. Reddy, MD, Mount Sinai Hospital, New York, New York, USA, presented results from the ongoing SELECT-LV study [NCT01905670], which showed that in patients with failed conventional cardiac resynchronization therapy (CRT; also called biventricular pacing), a wireless cardiac stimulation system provided synchronous left ventricular (LV) stimulation and significant reductions in QRS duration with no pericardial tamponade and few procedural adverse events.

SELECT-LV was designed to demonstrate the safe implantation of small receiver-electrodes into the endocardial surface of the left ventricle and to establish device utility in providing CRT in patients with heart failure. LV leadless pacing was performed via a 9-mm electrode implanted in the endocardial midlateral LV free wall. The electrode is activated by a submuscular ultrasonic transmitter synchronized to the right ventricular pacing pulse of a standard implantable cardioverter defibrillator pacemaker.

This was a nonrandomized, prospective, 6-center study in Europe in patients who met current guidelines for CRT but were considered CRT failures due to lead issues or nonresponse to CRT. The primary safety end point was safety during the 24-hour perioperative period, and the primary performance end point was safety and pacing at 1 month. Secondary end points at 6 months were safety, performance, and preliminary efficacy.

A total of 39 patients completed enrollment and screening. Implantation was attempted in 35 participants, 34 (97.1%) of which were successful. All implanted patients completed the 1-month follow-up; 26 patients completed the 6-month follow-up. Participants (88% men) were mean age 65.5 years with a mean ejection fraction (EF) of 26% and mean NYHA class of 2.6. At baseline, the mean intrinsic pre-implant QRS duration was 170 ms. About two-thirds of the patients were on anticoagulants prior to their implant. The most common reason for implantation was the inability to perform coronary sinus pacing.

There were a total of 4 procedure-related events in 4 patients (11.4%) during the < 24-hour perioperative period: groin fistula, which was surgically repaired; groin pseudoaneurysm, which required no treatment; ventricular fibrillation during electrode implant; and electrode embolization to the lower leg during catheter/dilator exchange (judged both procedure and device related). Unlike an earlier terminated study that reported a 30% tamponade rate [Auricchio A et al. Europace. 2014], this study had no episodes of pericardial tamponade.

During the 1-month follow-up period, there were 8 events in 8 patients (22.9%), including 1 death following prolonged resuscitation. Among these, there were 6 procedure-related events, including pseudoaneurysm, hematoma at transmitter pocket, a suspected infection, and a death following prolonged resuscitation. Furthermore, there was 1 procedure- and device-related event as well as atrial fibrillation (AF) episodes in a patient with AF, although these episodes were unrelated to the procedure or device. During the 6-month follow-up, there were 3 device-related events (premature battery depletion in 3 patients [11.5%]); these events were subsequently resolved. There were also 10 comorbidity events in 5 patients (19.2%), although most were unrelated to the procedure.

At 1 month, the primary end point, consistent CRT, was met in 97.1% of evaluable patients. At 1 month, the mean QRS reductions were 50 and 37 ms, as compared with baseline right ventricular and intrinsic pacing, respectively.

At 1 month, the primary performance end point of biventricular pacing capture on 12-lead electrocardiogram was met in 33 of 34 evaluable patients (97.1%). Furthermore, mean QRS duration with biventricular pacing was 50 ms shorter vs baseline right ventricular pacing and 37 ms shorter than baseline intrinsic pacing (Table 1).

EF increased by ≥ 5% in 69.5% of patients with a mean 6-month difference from baseline of 6.7%. NYHA decreased from 2.6 at baseline to 1.8 at 6 months (61.5% of patients had a ≥ 1 class improvement), and 64.7% of patients had ≥ 15% improvement in LV end-systolic volume.

The results of the SELECT-LV study demonstrated the feasibility of direct, wireless endocardial LV pacing for CRT. Overall CRT efficacy improved with evidence of significant reductions in QRS duration, synchronous LV stimulation, electrical and structural remodeling, and improvement in LVEF. Safety results improved compared with previous studies, but more data are needed.