Fellow MCS and Transplantation Houston, Texas, United States
Disclosure(s):
Maks Mihalj, MD, PhD: No financial relationships to disclose
Background: Impella devices provide rapid hemodynamic support in the management of cardiogenic shock but may disrupt aortic valve function. This study evaluated whether prior Impella exposure increases the risk of clinically significant aortic insufficiency (AI) after durable left ventricular assist device (LVAD) implantation.
Methods: We performed a single-center, retrospective cohort study of patients who received Impella support as a bridge to recovery or bridge to LVAD, or an LVAD implantation with no prior Impella bridging between January 2019 and August 2023, and investigated for AI progression over 2-year follow-up period. Two frameworks were used: the Impella Framework modeled the time from Impella implantation and compared AI progression between those who underwent Impella weaning to those bridged to the LVAD therapy. The LVAD Framework modeled the time from LVAD implantation and compared the AI progression between those who had prior bridging with Impella to those without. Patients with pre-existing moderate/severe AI or prosthetic valves were excluded. The primary endpoint was progression to moderate/severe AI or aortic valve replacement. Analyses included Kaplan–Meier survival curves and Fine–Gray competing risk models. Multivariable Cox regression was used to evaluate AI risk factors.
Outcome: During the study period, 315 patients received ≥1 Impella, and 167 underwent LVAD implantation. In the Impella Framework, AI progression was significantly higher in those bridged to LVAD therapy (n=59) compared to the Impella patients without subsequent LVAD therapy (n=256), with cumulative incidence of 36.4% versus 12.1% at 12 months and 46.8% versus 12.1% at 24 months (adjusted HR 3.27, 95% CI 1.57–6.82, p=0.002). In the LVAD Framework, AI progression was markedly higher in LVAD patients with prior Impella bridging (n=46) compared to those without prior Impella bridging (n=121), with 16.6% versus 0.9% at 12 months and 29.6% versus 6.0% at 24 months (adjusted HR 7.87, 95% CI 2.66–23.25, p< 0.001). Longer Impella support duration independently predicted AI progression (adjusted HR 1.01 per day; p=0.047). Mortality did not differ between LVAD groups.
Conclusion: Prior Impella support is strongly associated with increased risk of moderate/severe AI after LVAD-implantation, particularly with prolonged Impella runs. These findings underscore the need for heightened vigilance and may justify a lower threshold for concomitant AVR during LVAD-implantation in destination therapy. Prospective studies are warranted to guide future practice.
