A) At thermal equilibrium, when MR-active nuclei are placed in a magnetic field, spins that align parallel to the magnet field have a slightly lower energy than spins that align anti-parallel to it. The energy advantage obtained by aligning parallel to the magnetic field causes slightly more spins to point in that direction, i.e., causes a net polarization, which results in the production of MR signal. However, thermal equilibrium polarization is extremely low (on the order of only 0.0005%) and this directly translates into low MR signals. B) By forcing most spins to point in the same direction, DNP achieves polarization levels of > 20%, which translates into an equivalent 10,000-fold increase in MR signal. C) To achieve dynamic nuclear polarization (DNP), MR-active nuclei such as ¹³C are mixed with a low concentration of free electrons, and the sample is irradiated with microwaves in a high magnetic field (~3 T) and at low temperatures (~1 K). The hyperpolarizer system, shown on the left, also allows sample dissolution to temporarily maintain the high signal in solutions with a physiological temperature and pH for injection.

The metabolic pathways that have been studied using hyperpolarized ¹³C-labelled metabolic tracers and have the potential to be diagnostic targets for cardiovascular disease. LDH, lactate dehydrogenase; AAT, alanine aminotransferase; CA, carbonic anhydrase; CAT, carnitine acetyltransferase; GDH, glutamate dehydrogenase; PDH, pyruvate dehydrogenase.

A) The ¹³C-bicarbonate/[1-¹³C]pyruvate ratio in fed rats compared with fasted rats. The representative single spectra show that fasted rats had 74% lower ¹³C-bicarbonate production than their fed counterparts³³. Therefore, as would be predicted from previous in vitro studies of the glucose-fatty acid cycle, in vivo PDH flux is inhibited with fasting. B) Measurement of PDH flux based on in vivo hyperpolarized ¹³C MRS assessment of ¹³C-bicarbonate production significantly correlated with in vitro measurement of PDH activity. These data verified that hyperpolarized MRS measurement of ¹³C-bicarbonate production reflects in vivo PDH flux³⁴.

Representative spectrum acquired from a rat heart in vivo, after infusion of hyperpolarized [2-¹³C]pyruvate. For display purposes, 3 s of data were summed and a line broadening of 15 Hz was applied. This metabolic tracer has enabled dynamic, in vivo assessment of the first span of the Krebs’ Cycle. CAT, carnitine acetyltransferase; CS, citrate synthase; ICDH, isocitrate dehydrogenase; OMC, oxoglutarate-malate carrier; αKDH, α-ketoglutarate dehydrogenase.

A) Transverse proton magnetic resonance image of a mouse with a subcutaneously implanted tumor. B) Hyperpolarized ¹³C-bicarbonate was infused intravenously into the mouse and the infused ¹³C-bicarbonate was allowed to equilibrate with ¹³CO₂ based on the local extracellular tissue pH. A pH map was calculated across the tumor from the ratio of the ¹³C-bicarbonate and ¹³CO₂ voxel intensities acquired from ¹³C chemical shift images. C) and D) show the ¹³C-bicarbonate and ¹³CO₂ images, respectively.
In hearts, use of hyperpolarized ¹³C-bicarbonate to image ¹³C-bicarbonate and ¹³CO₂ to generate similar pH maps could be useful for detection of regions of ischemia. As in tumors, acidosis is known to occur in the ischemic myocardium. Reproduced with permission from⁵¹.

Proton images, semi-quantitative gadolinium perfusion maps and late gadolinium enhancement (LGE) images together with [1-¹³C]alanine and ¹³C-bicarbonate metabolic maps obtained in a pig after a 15 min and a 45 min occlusion of the left circumflex artery. The perfusion and metabolic maps have been superimposed on the proton image.
These images indicate the potential of hyperpolarized ¹³C MRI to demonstrate myocardial viability in patients with ischemic heart disease. Following the 15 min occlusion, the combination of preserved [1-¹³C]alanine production with a regional defect in ¹³C-bicarbonate production confirmed the presence of ‘stunned’ but viable myocardium (as determined by assessment of wall motion with proton cine-MRI). Following the 45 min occlusion, defects in both the [1-¹³C]alanine and ¹³C-bicarbonate metabolic maps correlated with a region of nonviable myocardium, as indicated by perfusion imaging and LGE. Adapted, with permission, from⁷⁶.

In vivo data acquired using a multi-slice cardiac-gated sequence showing spatial distribution of metabolites in a short-axis view of the healthy pig heart. Bicarbonate, lactate, and pyruvate volumes were acquired over 9 heartbeats, and the sequence was repeated for three time points. Data were acquired with a surface coil, resulting in higher signal detection from the anterior wall of the myocardium compared with the posterior wall. The resolution of the overlaid reconstructed images is 8.8 × 8.8 mm² in-plane with a 1 cm slice thickness. The entire scan was completed within 18 s.
The data acquired with this novel pulse sequence illustrate the increasingly high spatial and temporal resolution with which cardiac ¹³C metabolic images can be acquired. Further, images of pyruvate in the blood pool can be used for normalization of myocardial metabolite signals (i.e. bicarbonate). Reproduced with permission from¹¹⁴.