Related Experiment Video
Updated: Jan 3, 2026
Normothermic Ex Situ Heart Perfusion in Working Mode: Assessment of Cardiac Function and Metabolism
Published on: January 12, 2019
In vivo tissue reaction within the outflow conduit in patients supported by HeartWare HVAD
1Cardiology Department, St Vincent's Hospital, Sydney, Australia; University of New South Wales, Sydney, Australia.
Aim:
The frequency, extent, and nature of tissue ingrowth within the continuous-flow left ventricular assist device (cf-LVAD) outflow conduit has not been systematically assessed. We sought to characterize conduit histopathology at explantation in a cohort of patients with HeartWare ventricular assist device (HVAD) and assess the effect on pump performance.
Methods:
Patients undergoing routine histopathological assessment of a HeartWare HVAD removed at transplantation or autopsy were assessed. Outflow conduits were examined macroscopically, and visible tissue was sectioned for microscopic evaluation. In patients who had undergone prior contrast-enhanced computerized tomography (CT) with HVAD in situ, the outflow conduit was measured at the aortic anastomosis and 5 cm proximal to the anastomosis, in the axial and sagittal planes. All patients had their pump flow, flow pulsatility, current, and speed determined from log files examined at 1, 3, 6, 9, and 12 months after LVAD implantation.
Results:
Twenty-five consecutive patients were assessed (24 LVAD, 1 biventricular assist device (BiVAD)). Of the 26 outflow grafts assessed, there was evidence of tissue ingrowth reaction in 24 (92%) grafts. The most common site was the distal anastomosis (18/24, 75%), with the graft body involved in 14 of 24 (58%) grafts. Microscopic evaluation revealed acute inflammatory infiltrate in 4 of 24 grafts (17%), chronic inflammatory infiltrate in 14 of 24 (58%), neointima formation in 18 of 24 (75%) and fibrosis in 18 of 24 (75%) grafts. The median depth of tissue was 1 mm (range, 0-2 mm). The mean conduit diameter was 9.5 ± 0.6 mm at the aortic anastomosis compared with 11.1 ± 0.5 mm 5 cm proximal to the anastomosis (p < 0.0001). In patients with unchanged pump speed one month after implantation, analysis of log files revealed a significant (5.8 ± 8.6%) decrease in pump flow (4.65 ± 0.86 vs 4.38 ± 0.92 L/min, p = 0.01) and flow pulsatility (5.00 ± 1.10 vs 4.16 ± 1.05 L/min, p = 0.006).
Conclusions:
There is evidence of tissue formation within the HVAD outflow conduit in the vast majority of patients, most commonly located at the aortic anastomosis. This is associated with significantly decreased pump flow over time.
