Increasing evidence suggests that abnormal cardiac metabolism is a key causal factor in the development and worsening of several heart diseases. Cardiac metabolism has thus gained considerable attention worldwide lately, as it is believed to hold significant promises, both as a diagnostic and prognostication tool, as well as a novel target for treatment. Hyperpolarized MR has increased the sensitivity of MR >10.000 fold, thus allowing in situ interrogation of the pivotal changes in cardiac metabolism, which occur in conjunction with diseases[1]. In order to translate the promising findings in rodent models to human subjects it is essential to use large animal models resembling the human physiology, in the development of the needed cardiac protocols.As human trials involving hyperpolarized MR in the heart are imminent, we employed a clinically relevant, large animal model, and sought to evaluate the general feasibility of the method as well as its ability to detect an imposed shift in metabolic substrate utilization during metabolic modulation with glucose, insulin and potassium (GIK) infusion.Four overnight fasted healthy female Danish landrace pigs of weight 30kg were included in this study. The pigs were anaesthetized (Propofol (12mg initial dose, 0.4mg/kg/h thereafter for maintenance), fentanyl(8 µg/kg/h)), intubated and mechanically ventilated with a 60% O2-air mix. Catheterization was performed in both left and right femoral artery and vein for administration of hyperpolarized ¹³C-pyruvate, arterial blood pressure, blood glucose sampling and GIK infusion. To promote glucose metabolism, a continuous infusion of GIK (500ml 20% glucose, 50 IU/L insulin and potassium chloride 80mEq/L) was administered. The GIK infusion was started at 30min and kept constant at a rate of 90mL/h until the end of the last MRI scan. A clinical 3T GE HDx MR scanner (GE Healthcare, Milwaukee, WI, USA) was used to acquire ¹H images with an 8-channel cardiac array receiver coil (GE Healthcare, Milwaukee, WI, USA) and a ¹³C Helmholtz loop coil of 20cm diameter (PulseTeq Limited, Surrey, UK) was used for ¹³C-pyruvate examinations. The scan protocol was as follows: a scouting sequence to locate the heart, CINE left ventricular function (CINE-LVF) to measure ejection fraction (EF) prior to GIK, a Bloch-Siegert sequence to calibrate ¹³C power and frequency, a spiral CSI for ¹³C imaging and a CINE-LVF after the GIK infusion. The spiral CSI was positioned mid-ventricularly in the axial plane of the heart, was ECG- triggered in the diastolic phase and acquired during breath-hold. CINE-LVF was positioned to cover the left ventricle. The ¹³C CSI spiral images were acquired at 0min, 30min, 90min and 120min. Scan parameters for the sequences were: ¹³C CSI spiral, (8 repetitions, echo time (TE) 1.1ms, repetition time (TR) 100ms, flip angle (FA) 15°, matrix 60x60, field of view (FOV) 150x150mm2, in-plane resolution 2.5mm, slice thickness 50mm) and CINE-LVF (TE 1ms, TR 2.9ms, FA 35°, matrix 292x292, FOV 200x200mm2, in-plane resolution 0.7mm, slice thickness 8mm). Raw DICOM images were transferred to OsiriX (Pixmeo, Geneva, Switzerland) for anatomical overlay and region of interest (ROI) analysis. Statistics were performed with one-way ANOVA in Prism (GraphPad Software, Inc. La Jolla, CA, USA). A p-value below 0.05 was considered significant.Bicarbonate production was significantly increased at time point 120 min compared to 0, 30 and 90 min. Lactate production seemed to increase continuously during the course of each experiment. Alanine production showed no significant difference at any time point (Figure 1). Image quality of the ¹³C spiral CSI was consistently good and a representative time point during GIK infusion is shown in Figure 2 overlaid a CINE-LVF image at same position. EF pre and post was unaltered at 64±1% and arterial blood pressure, CO₂ and blood glucose levels during the experiments are shown in Figure 3. Blood glucose levels, starting at 4mmol/L and ending at 12mmol/L, showed expected response to the GIK infusion.In the normal heart, energy is primarily (>95%) produced via oxidative metabolism of free fatty acids (FFA) and glucose, with each fuel source contributing to the ATP-production with about 60-90% and 10-40%, respectively[2]. These ratios vary widely under normal physiological conditions; e.g. fasting increases reliance on FFA-metabolism greatly[3], but they also differ from the norm under pathological conditions. These metabolic shifts form the basis of interest when planning hyperpolarized MRI-studies of the heart, and it is of vital importance that the method is appropriate for their detection and possible quantification. This study demonstrates that hyperpolarized ¹³C-pyruvate, in a large animal, clinically relevant model, is a feasible method for cardiac studies, and, in combination with GIK intervention; that it is able to detect imposed metabolic shifts.