Pulsatility and flow patterns across macro- and microcirculatory arteries of continuous-flow left ventricular assist device patients

Eric J. Stöhr*, Ruiping Ji, Giulio Mondellini, Lorenzo Braghieri, Koichi Akiyama, Francesco Castagna, Alberto Pinsino, John R. Cockcroft, Ronald H. Silverman, Samuel Trocio, Oksana Zatvarska, Elisa Konofagou, Iason Apostolakis, Veli K. Topkara, Hiroo Takayama, Koji Takeda, Yoshifumi Naka, Nir Uriel, Melana Yuzefpolskaya, Joshua Z. WilleyBarry J. McDonnell, Paolo C. Colombo

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Background: Reduced arterial pulsatility in continuous-flow left ventricular assist devices (CF-LVAD) patients has been implicated in clinical complications. Consequently, recent improvements in clinical outcomes have been attributed to the "artificial pulse" technology inherent to the HeartMate3 (HM3) LVAD. However, the effect of the "artificial pulse" on arterial flow, transmission of pulsatility into the microcirculation and its association with LVAD pump parameters is not known. Methods: The local flow oscillation (pulsatility index, PI) of common carotid arteries (CCAs), middle cerebral arteries (MCAs) and central retinal arteries (CRAs-representing the microcirculation) were quantified by 2D-aligned, angle-corrected Doppler ultrasound in 148 participants: healthy controls, n = 32; heart failure (HF), n = 43; HeartMate II (HMII), n = 32; HM3, n = 41. Results: In HM3 patients, 2D-Doppler PI in beats with "artificial pulse" and beats with "continuous-flow" was similar to that of HMII patients across the macro- and microcirculation. Additionally, peak systolic velocity did not differ between HM3 and HMII patients. Transmission of PI into the microcirculation was higher in both HM3 (during the beats with "artificial pulse") and in HMII patients compared with HF patients. LVAD pump speed was inversely associated with microvascular PI in HMII and HM3 (HMII, r2 = 0.51, p < 0.0001; HM3 "continuous-flow," r2 = 0.32, p = 0.0009; HM3 "artificial pulse," r2 = 0.23, p = 0.007), while LVAD pump PI was only associated with microcirculatory PI in HMII patients. Conclusions: The "artificial pulse" of the HM3 is detectable in the macro- and microcirculation but without creating a significant alteration in PI compared with HMII patients. Increased transmission of pulsatility and the association between pump speed and PI in the microcirculation indicate that the future clinical care of HM3 patients may involve individualized pump settings according to the microcirculatory PI in specific end-organs.

Original languageEnglish
Pages (from-to)1223-1232
Number of pages10
JournalJournal of Heart and Lung Transplantation
Volume42
Issue number9
DOIs
Publication statusPublished - 10 Aug 2023

Keywords

  • HM3
  • HMII
  • LVAD
  • microcirculation
  • pulsatility

Cite this