Statistical shape modelling of the left ventricle for patients with HeartMate2 and HeartMate3 ventricular assist devices

Left ventricular assist devices (LVADs) have demonstrated promising outcomes in the management of end-stage heart failure. However, the altered intraventricular flow dynamics following LVAD implantation can lead to non-physiological shear rates and stagnant or recirculating zones, which increase the risk of thrombosis around the inflow cannula. There are conflicting recommendations regarding the optimal inflow cannula design and its association with thrombosis risk, possibly due to anatomical variations among patients. To explore the sources of these discrepancies, statistical shape models (SSMs) of the left ventricle (LV) were utilized to numerically evaluate the impact of anatomical variations on thrombosis risk. Nineteen CT scans of LVAD patients, consisting of 5 HeartMate2 (HM2) and 14 HeartMate3 (HM3) devices, were manually segmented. The coherent point drift (CPD) algorithm was implemented to register the segmented LVs. Separate SSMs were developed for HM2 and HM3 cohorts using a principal component analysis (PCA). Multiple anatomical metrics such as LV volume, sphericity, and cross-sectional circularity were compared. A computational fluid dynamics (CFD) analysis was performed for an end-stage heart failure condition characterised by rigid LV walls, closed aortic valve and LVAD flow rate of 5 L/min. Thrombosis risk was assessed by wall shear stress (WSS), stasis volume, turbulent kinetic energy (TKE) and washout. The HM2 and HM3 cohorts exhibited differences in sphericity, apical circularity, and conicity, which may be attributed to device shape and implantation technique. For both SSMs, larger LV volume was the main anatomical feature contributing to increased stasis volume and slower blood clearance, leading to higher thrombosis risk. The second anatomical metric contributing to increased thrombosis risk was reduced LV sphericity (HM2 patients) and formation of an apical bulge (HM3 patients). This study highlighted the statistical differences in LV shape between HM2 and HM3 patients, demonstrating how specific geometrical features of the LV may predispose patients to thrombus formation after LVAD implantation.

Authors

Institutions

Publication Details

Journal
Computers in Biology and Medicine
Published
2025-03-02
DOI
https://doi.org/10.1016/j.compbiomed.2025.109921
Citations
1
Primary Topic
Mechanical Circulatory Support Devices
Type
article
Field-Weighted Citation Impact
0.67

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Statistical shape modelling of the left ventricle for patients with HeartMate2 and HeartMate3 ventricular assist devices

Shaun D. Gregory, Andrew M. Taylor, Mehrdad Khamooshi, David C. McGiffin et al.
1 citations
Computers in Biology and Medicine
Mechanical Circulatory Support Devices
0.67
article

Statistical shape modelling of the left ventricle for patients with HeartMate2 and HeartMate3 ventricular assist devices

Shaun D. Gregory, Andrew M. Taylor, Mehrdad Khamooshi, David C. McGiffin, Diogo Ferreira de Almeida, Sam Liao, Marjan Azimi, M. Šeman
article en
1 citations

Abstract

Left ventricular assist devices (LVADs) have demonstrated promising outcomes in the management of end-stage heart failure. However, the altered intraventricular flow dynamics following LVAD implantation can lead to non-physiological shear rates and stagnant or recirculating zones, which increase the risk of thrombosis around the inflow cannula. There are conflicting recommendations regarding the optimal inflow cannula design and its association with thrombosis risk, possibly due to anatomical variations among patients. To explore the sources of these discrepancies, statistical shape models (SSMs) of the left ventricle (LV) were utilized to numerically evaluate the impact of anatomical variations on thrombosis risk. Nineteen CT scans of LVAD patients, consisting of 5 HeartMate2 (HM2) and 14 HeartMate3 (HM3) devices, were manually segmented. The coherent point drift (CPD) algorithm was implemented to register the segmented LVs. Separate SSMs were developed for HM2 and HM3 cohorts using a principal component analysis (PCA). Multiple anatomical metrics such as LV volume, sphericity, and cross-sectional circularity were compared. A computational fluid dynamics (CFD) analysis was performed for an end-stage heart failure condition characterised by rigid LV walls, closed aortic valve and LVAD flow rate of 5 L/min. Thrombosis risk was assessed by wall shear stress (WSS), stasis volume, turbulent kinetic energy (TKE) and washout. The HM2 and HM3 cohorts exhibited differences in sphericity, apical circularity, and conicity, which may be attributed to device shape and implantation technique. For both SSMs, larger LV volume was the main anatomical feature contributing to increased stasis volume and slower blood clearance, leading to higher thrombosis risk. The second anatomical metric contributing to increased thrombosis risk was reduced LV sphericity (HM2 patients) and formation of an apical bulge (HM3 patients). This study highlighted the statistical differences in LV shape between HM2 and HM3 patients, demonstrating how specific geometrical features of the LV may predispose patients to thrombus formation after LVAD implantation.

Computers in Biology and MedicineVol. 189
Queensland University of Technology (AU), The Alfred Hospital (AU), Department of Virology (FR), Monash Health (AU), Alfred Health (AU), Monash University (AU)
National Foundation for Medical Research and Innovation, National Heart Foundation of Australia, National Health and Medical Research Council
Good health and well-being
Openalex Percentile: Top 34%
Mechanical Circulatory Support Devices
0.67
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.