Background Current methods to quantify infarct size after myocardial infarction in mice are not ideal, requiring either tissue destruction for histology, or relying on non-direct measurements such as wall motion. High-field strength magnetic resonance imaging (MRI) can provide “virtual histology” and allows longitudinal studies of the same animal and, thereby, smaller group sizes. Therefore, we aimed to implement three-dimensional inversion–recovery late Gd enhancement MRI (3D-LGE), investigate Gd kinetics after intravenous (iv) and intraperitoneal (ip) injection and to validate the method against histology.
Methods Magnetic resonance experiments were carried out on a 9.4 Tesla system. A method to quantify myocardial T1 was developed and in-vivo reproducibility was tested. Then, using continuous T1 measurements, the Gd pharmacokinetics after iv and ip injection of 0.5 μmol/g were investigated in six healthy and six infarcted mice. Optimal imaging windows post-injection and inversion times yielding maximal viable/non-viable image contrast were determined for both injection routes. 3D-LGE infarct sizes were expressed as the percentage of left ventricular mass (%LV) and compared with tetrazolium-chloride histology in eight mice.
Results Baseline myocardial T1 was 952 ± 41 ms and deviation was less than 2% in repeated measurements (p = 0.43). T1 contrast between viable and non-viable myocardium was higher and peaked earlier after iv than after ip injection (see fig 1); however, both techniques retained a sufficient T1 contrast over a full 60 minutes. Infarct sizes derived from 3D-LGE and histology agreed well (36.0 ± 8.5%LV vs 35.1 ± 8.8%LV; r = 0.91, y = 0.948 + 0.0088, see fig 2), the mean difference of the methods was 1 ± 3.7%LV. Spatial resolution was 50 × 67 × 250 μm.
This study utilised two novel approaches to optimise the in-vivo assessment of myocardial infarction in mice. For the first time myocardial T1 maps were used to determine the T1 contrast between remote and infarcted myocardium after Gd injection over time and to predict suitable timing paramaters for late Gd enhancement imaging. This fully quantitative method found typical pharmacokinetics for both ip and iv injections. Thereby, optimal image contrast between viable and non-viable myocardium was ensured. The proposed approach enabled accurate infarct measurements with unsurpassed spatial resolution and remarkable tissue contrast in the mouse in vivo.
Implications 3D-LGE is suitable for replacing histological infarct measurements in the mouse in vivo. This is particularly advantageous when infarct measurements before intervention and preservation of myocardial tissue are required or when longitudinal assessment of the same mouse is desirable.