Pilot study on the feasibility of replacing invasive heart pressure measurements with non-invasivemagnetic resonance elastography as a way to reduce rodent numbers in pre-clinical research

Abstract

Background: Magnetic Resonance Imaging (MRI) and Spectroscopy (MRS) are phenotyping techniques that allow for a non-invasive characterization of cardiac function in rodents. However, when assessing myocardial function, the isovolumic phases of the cardiac cycle are not directly accessible to imaging techniques, yet this is where maximal rates of pressure generation and relaxation occur. This type of information is only available using LV catheterization, which in rodents is an invasive and terminal procedure, requiring new experimental groups every time-point. Magnetic Resonance Elastography (MRE) measures shear deformation following mechanical tissue stimulation and has the potential to non-invasively access ventricular pressure in vivo, as successfully demonstrated in pig and human hearts. Therefore, this project aims to build a setup, which allows for mechanical stimulations inside the magnet with the appropriate amplitude and frequency range required for murine experiments. Methodology: Core to this project was the development of the MRE hardware required for the mechanical tissue stimulation and compatible with the existing animal handling system. MRI sequences needed to be implemented to equip them with motion-sensitizing gradient to encode shear deformation. In order to validate the MRE setup and the MRI sequences, phantoms of different concentrations of agarose were excited over a wide range of frequencies. The frequencies ranged from hundreds of Hz up to 1.5 kHz. MRE experiments were subsequently performed on a two-compartment phantom with different concentrations of agarose. Finally, a dead mouse was subjected to MRE examinations. Results: There was a clear uniform penetration of shear waves in phantom experiments. All phantoms demonstrated a pattern of phase shifts across the sample, which corresponded to propagation of mechanical vibrations. The gel of 0.5% agarose deformed more than the 1% agarose gel, resulting in a wavelength shorter than the wavelength found in stiffer gel. The results also show higher attenuation of shear waves in 1% agarose gel, either at 500 Hz or 1000 Hz of vibration frequency, when compared with the propagation of waves in 0.5% agarose gel. In the two-compartment phantom, shear waves penetrated non-uniformly the phantom. There was a better propagation of waves in the 0.5% gel than in the 2% one, resulting in a shorter wavelength in the softer gel than the rigid one. In the gel with higher level of stiffness the wavelength exceeded the dimensions of the gel while in the softer gel were observed six complete vibration cycles. In the mouse experiments, there was a strong tissue distinction when vibrations passed through them. The tissues deformed depending on their rigidity which translated into different spins phase variations either in the liver or in the heart. Conclusions: A setup to provide tissue mechanical stimulation in a small bore pre-clinical MR system was successfully developed. It was qualitatively demonstrated that the application of shear wave mechanical excitation may be a method to encode mechanical properties of a heterogeneous object. Thus, this project represented a major step in the ability to develop a technique for indirect LV pressure measurements in mice.

Date of Award	2013
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Jürgen Schneider (Supervisor) & Mahon Maguire (Co-Supervisor)