Development and validation of a computational fluid dynamics methodology for simulation of pulsatile left ventricular assist devices

Richard B. Medvitz, James W. Kreider, Keefe B. Manning, Arnold Anthony Fontaine, Steven Deutsch, Eric G. Paterson

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

An unsteady computational fluid dynamic methodology was developed so that design analyses could be undertaken for devices such as the 50cc Penn State positive-displacement left ventricular assist device (LVAD). The piston motion observed in vitro was modeled, yielding the physiologic flow waveform observed during pulsatile experiments. Valve closure was modeled numerically by locally increasing fluid viscosity during the closed phase. Computational geometry contained Bjork-Shiley Monostrut mechanical heart valves in mitral and aortic positions. Cases for computational analysis included LVAD operation under steady-flow and pulsatile-flow conditions. Computations were validated by comparing simulation results with previously obtained in vitro particle image velocimetry (PIV) measurements. The steady portion of the analysis studied effects of mitral valve orientation, comparing the computational results with in vitro data obtained from mock circulatory loop experiments. The velocity field showed good qualitative agreement with the in vitro PIV data. The pulsatile flow simulations modeled the unsteady flow phenomena associated with a positive-displacement LVAD operating through several beat cycles. Flow velocity gradients allowed computation of the scalar wall strain rate, an important factor for determining hemodynamics of the device. Velocity magnitude contours compared well with PIV data throughout the cycle. Computational wall shear rates over the pulsatile cycle were found to be in the same range as wall shear rates observed in vitro.

Original languageEnglish (US)
Pages (from-to)122-131
Number of pages10
JournalASAIO Journal
Volume53
Issue number2
DOIs
StatePublished - Mar 1 2007

Fingerprint

Left ventricular assist devices
Heart-Assist Devices
Hydrodynamics
Velocity measurement
Pulsatile flow
Computational fluid dynamics
Rheology
Shear deformation
Pulsatile Flow
Computational geometry
Flow simulation
Hemodynamics
Steady flow
Unsteady flow
Flow velocity
Pistons
Strain rate
Equipment and Supplies
Experiments
Viscosity

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Bioengineering
  • Biomaterials
  • Biomedical Engineering

Cite this

@article{726cae550a1d4375888de68ecbdaa1f3,
title = "Development and validation of a computational fluid dynamics methodology for simulation of pulsatile left ventricular assist devices",
abstract = "An unsteady computational fluid dynamic methodology was developed so that design analyses could be undertaken for devices such as the 50cc Penn State positive-displacement left ventricular assist device (LVAD). The piston motion observed in vitro was modeled, yielding the physiologic flow waveform observed during pulsatile experiments. Valve closure was modeled numerically by locally increasing fluid viscosity during the closed phase. Computational geometry contained Bjork-Shiley Monostrut mechanical heart valves in mitral and aortic positions. Cases for computational analysis included LVAD operation under steady-flow and pulsatile-flow conditions. Computations were validated by comparing simulation results with previously obtained in vitro particle image velocimetry (PIV) measurements. The steady portion of the analysis studied effects of mitral valve orientation, comparing the computational results with in vitro data obtained from mock circulatory loop experiments. The velocity field showed good qualitative agreement with the in vitro PIV data. The pulsatile flow simulations modeled the unsteady flow phenomena associated with a positive-displacement LVAD operating through several beat cycles. Flow velocity gradients allowed computation of the scalar wall strain rate, an important factor for determining hemodynamics of the device. Velocity magnitude contours compared well with PIV data throughout the cycle. Computational wall shear rates over the pulsatile cycle were found to be in the same range as wall shear rates observed in vitro.",
author = "Medvitz, {Richard B.} and Kreider, {James W.} and Manning, {Keefe B.} and Fontaine, {Arnold Anthony} and Steven Deutsch and Paterson, {Eric G.}",
year = "2007",
month = "3",
day = "1",
doi = "10.1097/MAT.0b013e31802f37dd",
language = "English (US)",
volume = "53",
pages = "122--131",
journal = "ASAIO Journal",
issn = "1058-2916",
publisher = "Lippincott Williams and Wilkins",
number = "2",

}

Development and validation of a computational fluid dynamics methodology for simulation of pulsatile left ventricular assist devices. / Medvitz, Richard B.; Kreider, James W.; Manning, Keefe B.; Fontaine, Arnold Anthony; Deutsch, Steven; Paterson, Eric G.

In: ASAIO Journal, Vol. 53, No. 2, 01.03.2007, p. 122-131.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development and validation of a computational fluid dynamics methodology for simulation of pulsatile left ventricular assist devices

AU - Medvitz, Richard B.

AU - Kreider, James W.

AU - Manning, Keefe B.

AU - Fontaine, Arnold Anthony

AU - Deutsch, Steven

AU - Paterson, Eric G.

PY - 2007/3/1

Y1 - 2007/3/1

N2 - An unsteady computational fluid dynamic methodology was developed so that design analyses could be undertaken for devices such as the 50cc Penn State positive-displacement left ventricular assist device (LVAD). The piston motion observed in vitro was modeled, yielding the physiologic flow waveform observed during pulsatile experiments. Valve closure was modeled numerically by locally increasing fluid viscosity during the closed phase. Computational geometry contained Bjork-Shiley Monostrut mechanical heart valves in mitral and aortic positions. Cases for computational analysis included LVAD operation under steady-flow and pulsatile-flow conditions. Computations were validated by comparing simulation results with previously obtained in vitro particle image velocimetry (PIV) measurements. The steady portion of the analysis studied effects of mitral valve orientation, comparing the computational results with in vitro data obtained from mock circulatory loop experiments. The velocity field showed good qualitative agreement with the in vitro PIV data. The pulsatile flow simulations modeled the unsteady flow phenomena associated with a positive-displacement LVAD operating through several beat cycles. Flow velocity gradients allowed computation of the scalar wall strain rate, an important factor for determining hemodynamics of the device. Velocity magnitude contours compared well with PIV data throughout the cycle. Computational wall shear rates over the pulsatile cycle were found to be in the same range as wall shear rates observed in vitro.

AB - An unsteady computational fluid dynamic methodology was developed so that design analyses could be undertaken for devices such as the 50cc Penn State positive-displacement left ventricular assist device (LVAD). The piston motion observed in vitro was modeled, yielding the physiologic flow waveform observed during pulsatile experiments. Valve closure was modeled numerically by locally increasing fluid viscosity during the closed phase. Computational geometry contained Bjork-Shiley Monostrut mechanical heart valves in mitral and aortic positions. Cases for computational analysis included LVAD operation under steady-flow and pulsatile-flow conditions. Computations were validated by comparing simulation results with previously obtained in vitro particle image velocimetry (PIV) measurements. The steady portion of the analysis studied effects of mitral valve orientation, comparing the computational results with in vitro data obtained from mock circulatory loop experiments. The velocity field showed good qualitative agreement with the in vitro PIV data. The pulsatile flow simulations modeled the unsteady flow phenomena associated with a positive-displacement LVAD operating through several beat cycles. Flow velocity gradients allowed computation of the scalar wall strain rate, an important factor for determining hemodynamics of the device. Velocity magnitude contours compared well with PIV data throughout the cycle. Computational wall shear rates over the pulsatile cycle were found to be in the same range as wall shear rates observed in vitro.

UR - http://www.scopus.com/inward/record.url?scp=34247146434&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34247146434&partnerID=8YFLogxK

U2 - 10.1097/MAT.0b013e31802f37dd

DO - 10.1097/MAT.0b013e31802f37dd

M3 - Article

C2 - 17413548

AN - SCOPUS:34247146434

VL - 53

SP - 122

EP - 131

JO - ASAIO Journal

JF - ASAIO Journal

SN - 1058-2916

IS - 2

ER -