PET/MRI assessment of the infarcted mouse heart

Abstract

Heart failure originating from myocardial infarction (MI) is a leading cause of death worldwide. Mouse models of ischaemia and reperfusion injury (I/R) are used to study the effects of novel treatment strategies targeting MI, however staging disease and treatment efficacy is a challenge. Damage and recovery can be assessed on the cellular, tissue or whole-organ scale but these are rarely measured in concert. Here, for the first time, we present data showing measures of injury in infarcted mice using complementary techniques for multi-modal characterisation of the heart. We use in vivo magnetic resonance imaging (MRI) to assess heart function with cine-MRI, hindered perfusion with late gadolinium enhancement imaging and muscular function with displacement encoded with stimulated echoes (DENSE) MRI. These measures are followed by positron emission tomography (PET) with 18-F-fluorodeoxyglucose to assess cellular metabolism. We demonstrate a protocol combining each of these measures for the same animal in the same imaging session and compare how the different markers can be used to quantify cardiac recovery on different scales following injury.

Keywords: PET/MRI; mouse; multimodality; myocardial infarction.

Figure 1

Overview of our sequential multi-modality imaging protocol. The time spent in the MRI and the time spent in the PET are similar, suggesting that the same examination performed on a combined PET/MRI scanner would take half the time.

Figure 2

Ejection fraction, DENSE-derived strain and PET infarct size plotted against LGE MRI infarct size.

Figure 3

End systolic short-axis slice from a single mouse: a) voxel-wise displacements measured by DENSE MRI, areas of reduced contraction are indicated in red. b) radial strain. c) circumferential strain: areas of increased circumferential strain correspond to the hypokinetic areas in the displacement map.

Figure 4

One short-axis slice from a single mouse. a) End diastolic LGE image, areas of hyperenhancement correspond to non-viable tissue. b) End systolic DENSE-derived Displacement map: a hypokinetic area (marked in red) is present larger than the infarct. c)End diastolic FDG-PET uptake is reduced in the infarcted areas, although small infarcts are not visualised in the PET image.

Figure 5

Complementary information from multi-modality imaging can provide an accurate assessment of different aspects of a putative treatment: a) LGE MRI shows non-viable tissue. b) Cine MRI evaluates global heart function. c) DENSE MRI enquires local muscle performance. d) PET is used for molecular imaging.