Survival after Heart Transplantation in LVAD-Supported Patients: Impact of Implantation Technique and Duration of Support

Objectives: Postcardiac transplant outcome in patients with pulsatile left ventricular assist devices (LVAD) has been described to be significantly impaired. There are still controversial reports regarding the results after heart transplantation (htx) in patients previously supplied with continuous-flow devices. We aimed to analyze our patients particularly with regard to LVAD implantation technique and duration of pretransplant support.

Methods: Between 2010 and July 2018, 108 patients underwent htx in our department. Sixty-nine of these patients were supported by an LVAD (HeartMate II, HeartMate 3, or HeartWare HVAD) before transplant (group VAD). In 10 out of those patients, VAD was implanted in a minimally invasive technique (partial upper sternotomy + left minithoracotomy) (group MIC). Thirty-nine patients without VAD before htx served as controls. VAD and non-VAD recipients were comparable regarding further preoperative status, including high-urgent percentage. Additionally, we did not find any donor-related differences between the groups.

Results: A 30-day mortality was 15.9% in group VAD (MIC: 11.1%) compared with 5.1% in controls (p > 0.05). One-year survival was significantly higher in non-VAD patients with 85.7% compared with 66.7%.

Posttransplant morbidity was overall increased in VAD patients, significantly only for severe infections with 18.8% compared with 5.3%. Postoperative need for va-ECMO (33.3 vs. 17.9%), neurologic complications (17.4 vs. 7.9%), and resternotomy due to bleeding (30.4 vs. 7.9%) occurred more often in VAD recipients, however, each just without significance. In patients with MIC-VAD implantation, there was no need for resternotomy after htx, meaning a significant difference to all other VAD recipients and even to controls.

Duration of mechanical ventilation was comparable between the groups, whereas stay on intensive care unit and in hospital was significantly prolonged in VAD recipients.

Together with a slightly higher morbidity, patients with a pretransplant LVAD support > 2 years displayed a reduced 1-year survival.

Conclusion: With our single-center experience, we could show worse results after htx in patients with pretransplant LVAD support. Despite mostly missing significance also continuous-flow devices definitely impair outcome after htx, particularly if LVAD support exceeded 2 years, likely caused by HLA sensitization. Presumably due to less adhesions, a minimally invasive LVAD implantation significantly reduced posttransplant bleeding complications.