Real-world long-term one-centre experience with the use of 113 fully magnetically levitated continuous flow left ventricular assist devices

Abstract Introduction Owing to increasing numbers of heart failure (HF) patients (pts) the need for left ventricular assist device (LVAD) expands. Although this therapy improves survival in severe HF pts it is not free from limitations. Background The purpose of this study was to analyze outcomes of fully magnetically levitated LVAD implantations in our institution. Methods We retrospectively analyzed data of all consecutive 113 HeartMate3 LVAD pts (90% male; mean age-56 y; mean BMI- 28.1; median INTERMACS profile -3.1, other patient characteristics depicted in Table 1) implanted in our institution within years 2016-2024. The mean time of LVAD support was 833 days (median 619, range 1-2837). The probability of survival (Kaplan-Meier) was 0.88; 0.77; 0.69; 0.54; 0.4; 0.31 and 0.23 for 1,6,12,24,36,48, 60 months respectively (Figure 1). Patients were followed to death, heart transplantation, LVAD explantation or to the end of observation in our institution. 26 pts (23%) were transplanted, 52(46%) died during LVAD support and no pumps were explanted or de-activated. Results Early right ventricular failure (RVF) occurred in 32 (28% ) of pts, while late RVF only in 9 (8%). Right ventricular assist device (RVAD) was used in 10 pts(9%); concomitant valvular surgery was performed in 16 pts(14%). Drive-line infection (DLI), defined as at least one positive wound culture, was found in 47 pts(42%), while recurrent DLI in 36 pts( 32%). At least one positive blood culture during LVAD support occurred in 34 pts(30%). Ischemic stroke (IS) affected 11 pts(10%), hemorrhagic stroke (HS) – 7 pts(6%), gastrointestinal bleeding (GIB) - 13 pts(11%), pump thrombosis - 1 patient, outflow graft obstruction (OGO) - 3 pts(2.6% ). Mean time to death was 484 days (median 202, range 1-2446), while time to first positive drive-line wound culture - 571 (median 452, range 11-2043), time to first positive blood culture- 362 (median 41, range 5-2504), to IS- 82 (median 1 day, range 0-830); HS- 693 (median 449, range 5-2444), GIB- 297 (median 49, range 3-1227). We found statistically significant correlations (by use of log-rank test) between death during LVAD support and ischemic HF, HS, GIB, early and late RVF, RVAD use, DLI or recurrent DLI ( p respectively: 0.012, 0.019, 0.044, 0.006, 0.009, 0.001, 0.033, 0.01). No statistically significant relations were found between death and non-ischemic HF, IS, positive blood culture during LVAD support and concomitant valvular procedure at LVAD implantation ( p respectively: 0.72, 0.57, 0.49, 0.074). Conclusions Despite evident progress of LVAD support outcomes and significant reduction of hemocompatibility related events with fully magnetically levitated pumps, DLI and early RVF still remain major complications while hemorrhagic adverse events have a negative impact on survival of LVAD recipients. Further research is needed to achieve improvement in this area including establishment of optimal antithrombotic therapy and device innovations.Figure 1 Table 1