CARC9: Outcomes of HeartMate 2 and HVAD Exchange to HeartMate 3 Left Ventricular Assist Device

Background: Device exchange may be required after left ventricular assist device (LVAD) implantation in the setting of pump malfunction or thrombosis. We evaluated clinical outcomes following device exchange to a HeartMate 3 (HM3) from two different pump models to assist in the surgical and medical decision-making process. Methods: We retrospectively assessed all patients (n=14) who underwent a device exchange from a HeartMate 2 (HM2, n = 7) or HVAD (n = 7) to a HM 3 at our institution. The primary outcome was survival at 6 months, a secondary analysis evaluated postoperative outcomes through discharge disposition. Baseline characteristics and clinical results were compared using the Fisher’s exact and Mann-Whitney U tests. Results: Of the 14 destination therapy patients, 2 from the HM2 to HM3 group had previously undergone a HM2 exchange and 1 HVAD to HM3 patient had two prior HVAD exchanges. Baseline characteristics were similar between the two cohorts (Table). The HM2 patients had a significantly longer duration of device support (4.4 vs. 2.5 years, p =.004) and less thrombosis (57% vs. 100%, p = 0.19) compared to the HVAD group. The HVAD patients had a higher international normalized ratio (2.1 vs. 1.1, p =.06) and trend towards a lower INTERMACS profile 1-3 (43% vs. 14%, p =.56) on day of exchange compared to the HM2 group. The inflow cannula of the HM3 was placed into the HM2 sewing ring, and an end-to-end anastomosis was performed from the HM3 outflow graft to the proximal aspect of the remaining portion of HM2 outflow graft for all HM2 to HM3 exchanges. Except for one patient, all HVAD pump components including the sewing ring, were replaced during device exchange. The 30-day and 6-month survival in the HM2 to HM3 vs. HVAD to HM3 cohorts was 100% vs. 71%, p =.43 and 100% vs. 57%, p =.19, respectively. Postoperatively, the HVAD to HM3 group had significantly more right heart failure requiring RVAD support (71% vs. 0%, p =.02), ventilator days (11 vs. 1, p =.001), and required reoperation more often (100% vs. 14%, p =.004) than the HM2 to HM3 group. The HVAD to HM3 patients also experienced greater time on inotropes, major bleeding and infection events, renal failure requiring dialysis, and a longer hospital stay (Table). Of the 7 HM2 to HM3 patients, 5 (71%) were discharged to home and 2 (29%) to acute inpatient rehab. No HVAD to HM3 patients were discharged directly to home. There were 2 (29%) in-hospital deaths, 1 (14%) long-term acute care facility, and 4 (57%) required acute inpatient rehab. Conclusion: For patients undergoing a device exchange from HVAD to HM3, this study suggests a higher incidence of postoperative complications including right ventricular failure requiring RVAD support, prolonged ventilator support, and an increased need for reoperation with a trend towards lower 6-month survival compared to HM2 to HM3 patients.