Author + information
- Giovanni Donato Aquaro, MD∗ (, )
- Francesca Frijia, MSc,
- Vincenzo Positano, MSc,
- Luca Menichetti, PhD,
- Maria Filomena Santarelli, PhD,
- Jan Henrik Ardenkjaer-Larsen, PhD,
- Florian Wiesinger, PhD,
- Vincenzo Lionetti, MD, PhD,
- Simone Lorenzo Romano, MD,
- Giacomo Bianchi, MD,
- Danilo Neglia, MD,
- Giulio Giovannetti, MD,
- Rolf F. Schulte, PhD,
- Fabio Anastasio Recchia, MD, PhD,
- Luigi Landini, PhD and
- Massimo Lombardi, MD
- ↵∗Fondazione G. Monasterio CNR-Regione Toscana, Via G. Moruzzi, 1, 56124, Pisa, Italy.
The objective of this study was to evaluate the capability and accuracy of cardiac magnetic resonance with hyperpolarized [1-13C]-pyruvate using the fast 3-dimensional (3D) pulse sequence to detect the presence and regional distribution of transient cardiac metabolic changes in a pig model of ischemia–reperfusion.
In 7 male pigs, a pneumatic coronary occluder was placed around the left anterior descending coronary artery. [1-13C]-pyruvate polarization was performed using dynamic nuclear polarization as previously described (1). Injections were performed at rest, during coronary occlusion, and during reperfusion. A 3D IDEAL spiral sequence was used at 3-T cardiac magnetic resonance (2). Metabolite signal was evaluated in 120 myocardial sectors. The metabolic activity mismatch (MAM) between 2 segmental variation maps was defined as 100(Sai−Sbi)/(Sai+Sbi)/2, where Sai and Sbi are the relative values of the signal of metabolite in the segment “i” in condition “a” and “b”.
The MAM of lactate and bicarbonate of the ischemic segments (middle and apical anteroseptal and anterior segments) was significantly different from that of the remote regions. In reperfusion, a significant inhomogeneity of MAM of bicarbonate (p < 0.001) was found, whereas no significant difference was found for lactate. Figure 1 shows a different distribution for lactate (−21 ± 6 vs. 3 ± 5, p < 0.001) or bicarbonate metabolic activity (−29 ± 7 vs. 33 ± 6, p < 0.0001) in the left ventricular segments involved in the ischemic process than in the remote segments. In reperfusion, lactate signal increased (20 ± 10 vs. −7 ± 5, p = 0.007) and bicarbonate decreased (−38 ± 12 vs. 36 ± 11, p < 0.0001) in the involved segments.
We evaluated cardiac metabolism in vivo during ischemia and acute reperfusion with a whole-heart acquisition using volume coils and high spatial resolution acquisition. Neither of these requirements was met by previously described methods. The main finding of the current study was that spatial resolution obtained using a 3D IDEAL spiral chemical shift imaging pulse sequence was high enough to provide 3D information on acute changes in pyruvate and metabolites in left ventricular myocardium using the conventional regional segmentation.
The IDEAL spiral chemical shift imaging pulse sequence allows the complete 3D dataset of information to be obtained simultaneously for each metabolite, with optimal signal-to-noise ratio and short acquisition time. This might be relevant for the application of hyperpolarized [1-13C]-pyruvate in humans, allowing complete acquisition during 1 breath-hold, increasing the signal-to-noise ratio and minimizing the effect of the main limitation of this technique, which is the fast signal decay of hyperpolarized substrates.
- American College of Cardiology Foundation