Abstract 16561: Noninvasive Left Ventricle Pressure-Volume Loop Determination Method With Cardiac Magnetic Resonance Imaging and Carotid Tonometry Using a Physics-Informed Approach
Jing Liu, Alda Bregasi, Gary F Mitchell, Niema M Pahlevan- Physiology (medical)
- Cardiology and Cardiovascular Medicine
Introduction: The left ventricle (LV) pressure-volume loop (PV-loop) is a powerful tool for evaluating LV function. Current non-invasive cardiac MRI-based methods for estimating LV PV-loops provide limited diastolic LVP information (Circ Cardiovasc Imaging. 2019; PMID: 30630347).
Aim: In this study, we introduced a new noninvasive method for PV-loop determination from cardiac MRI and noninvasive carotid pressure waveform using a physics-informed optimization method.
Methods: LV volumes (LVV) were computed using short axis MRI images from a standard clinical protocol, while noninvasive carotid pressure waveforms were obtained using arterial tonometry in a clinical cohort consisting of 51 healthy and 25 heart failure (HF) participants. We formulated constraints and objective functions to optimize a time-varying LV elastance model (double-Hill) based on LVV and carotid pressure waveforms, considering LV-arterial coupling during aortic valve opening and decoupling during closure. This model allowed for individualized characterization of systolic and diastolic LV pressure and LV PV-loop.
Results: In HF, our PV-loop analysis produced significantly smaller stroke work (p<0.0001), mean external power (p<0.05), ESPVR (p<0.0001), ventricular efficiency (p<0.0001), Ees/Ea (p<0.0001), LV systolic pressure (p<0.05), and +dP/dt (p<0.05. Additionally, there was an increase in potential energy (p<0.0001) and LV end-diastolic pressure (LVEDP) (p<0.05) for the HF group (Fig. 1). These observations were all consistent with general PV-loop characteristics of HF patients. Previous methods failed to capture certain functional characteristics of HF, including reduced stroke work, decreased mean external power, and elevated LVEDP.
Conclusions: Our physics-informed optimization method effectively captures HF-related functional changes in LV PV-loops. This method may facilitate noninvasive HF diagnostics and lead to improved therapeutic management.