Myocardial metabolic flexibility following ketone infusion demonstrated by hyperpolarized [2-13C]pyruvate MRS in pigs | Scientific Reports

Scientific Reports volume 15, Article number: 5849 (2025) Cite this article

286 Accesses

3 Altmetric

Metrics details

This study aims to investigate the effects of β-3-hydroxybutyrate (β-3-OHB) infusion on myocardial metabolic flexibility using hyperpolarized [2-13C]pyruvate magnetic resonance spectroscopy (MRS) in the pig heart. We hypothesized that β-3-OHB infusion will cause rapid, quantifiable alterations in tricarboxylic acid (TCA) cycle flux as measured non-invasively by 13C MRS and reflect myocardial work. Five female Danish landrace pigs underwent β-3-OHB infusion during a hyperinsulinemic euglycemic clamp (HEC). Cardiac metabolism and hemodynamics were monitored using hyperpolarized [2-13C]pyruvate MRS and cardiac MRI. β-3-OHB infusion during HEC resulted in significant increases in cardiac output over baseline (from 1.9 to 3.8 L/min, p = 0.0011) and heart rate (from 51 to 85 bpm, p = 0.0004). Metabolic analysis showed a shift towards increased lactate production and decreased levels of acetyl-carnitine and glutamate during β-3-OHB infusion. Following the termination of the infusion, a normalization of these metabolic markers was observed. These results demonstrate the profound metabolic adaptability of the myocardium to ketone body utilization. The infusion of Na-β-3-OHB significantly alters both the hemodynamics and metabolism of the porcine heart. The observed increase in cardiac output and metabolic shifts towards lactate production suggest that ketone bodies could potentially enhance cardiac function by providing an efficient-energy substrate that, if provided, is preferentially used. This study provides new insights into the metabolic flexibility of the heart and hints at the potential therapeutic benefits of ketone interventions in heart failure treatment.

Ketones bodies (KBs) have received much recent attention as an exogeneous novel dietary supplement, with effects on both athletic performance and potentially in the treatment of disease1. Ordinarily present at low concentrations, KBs have been found to play an important role in cardiac metabolism: (1) they are elevated in the plasma of heart failure patients2; (2) may be implicated in the potential mediators of the cardioprotective effects of SGLT-2 inhibitors such as empagliflozin3,4,5,6,7,8; and (3) are of particular interest owing to a fundamental advantage in oxygen efficiency, in terms of moles of ATP produced per mole of oxygen required9. Evolutionarily, KBs are only present in mammalian adults in cases of extreme starvation (or diabetic crisis) during which they are utilized in preference to other substrates if available, which can result in energetic benefits during metabolic stress10. Interestingly, KBs are subject to substantially fewer points of regulation in comparison to lipids or sugars: for example, β-hydroxybutyrate dehydrogenase is constitutively expressed in mouse liver and brain11, and the flux through it primarily is determined by mass action12, in contrast to the multiple sites of control and regulation within the (more complex) pathways representing fatty-acid oxidation or glycolysis.

In heart failure, both the uptake and oxidation of substrates such as free fatty acids and glucose are impaired, with a substrate selection shift towards glycolysis over beta oxidation13,14,15,16, while ketone utilization is maintained2. This has led to the investigation of ketone body utilization as an adjunct therapy using either oral or intravenous supplementation with the ketone body, β-3-hydroxybyturate (β-3-OHB), which is endogenously produced by the liver under conditions of prolonged fasting or ketogenic diets. In these situations, β-3-OHB is hepatically exported as a fuel source, circulates to peripheral tissues and is both readily transported, converted to acetoacetate (AcAc) and utilized as a metabolic fuel. The exogeneous supplementation with ketone bodies whilst not necessarily under ketogenic conditions has shown promising hemodynamic effects in both healthy individuals and patients with heart failure, where an increase in myocardial blood flow, heart rate, stroke volume and cardiac output (CO) has been observed17,18,19,20. However, the mechanism behind these effects is still not fully understood.

Hyperpolarized magnetic resonance spectroscopy (MRS) enables evaluation of the local metabolic status in the heart tissue9, thus providing a unique opportunity to explore the effects of different factors, such as β-3-OHB, on heart metabolism and function. MRS with hyperpolarized pyruvate permits the determination of the tricarboxylic acid (TCA) cycle flux in real time and shift in substrates utilization18,21,22, making it possible to investigate the effects of ketones on the cardiac metabolism in vivo in pigs, commonly used as a large animal model in cardiovascular research.

This study therefore aims to investigate the effects of an infusion of the ketone body (sodium) β-3-OHB, on metabolic and hemodynamics changes in the heart during a hyperinsulinemic euglycemic glucose clamp through hyperpolarized [2-13C]pyruvate MRS. We hypothesize that the resulting rapid alterations in TCA cycle flux would be directly quantifiable and reflective of myocardial work and haemodynamic parameters.

Five non-fasted female Danish landrace pigs (originally crossed from Landrace, Duroc and Yorkshire pigs, 38–42 kg body weight, targeting a nominal weight of 40 ± 5 kg pigs) were used in this experiment, as described in detail in previous work quantifying myocardial metabolic alterations following glucose, insulin and potassium infusion23. Pigs were born in a conventional sow herd in Denmark, located in the Silkeborg municipality, in the central region of Denmark. The farm is under the jurisdiction of the Danish Veterinary and Food Administration. Pigs were anaesthetised and observed throughout 3 h in the MR scanner during the experiment, and all animal experiments were undertaken following appropriate local ethical review, with the ARRIVE guidelines, and in accordance with the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals and Danish law, regulations and guidelines and were approved by the Danish Animal Experiments Inspectorate under the Danish Veterinary and Food. Anaesthesia was induced and maintained with continuous intravenous infusions of propofol (Fresenous Propolipid 2% emulsion; 3.75 mg/kg/h dose; ~ 15 ml/h) and fentanyl (~ 0.03 mg/kg/h), intubated and mechanically ventilated utilising a commercially available Avance respirator system with a built-in CO2 and O2 gas monitoring unit (GE Healthcare, Broendby, Denmark). Three 6 F introducer sheaths were established in the femoral artery and veins by means of an ultrasound-guided Seldinger technique, permitting arterial blood pressure measurement, blood sampling and infusion delivery. A solution of 15% w/v (Na)-β-3-OHB (D/L-β-hydroxybutyric acid, Sigma-Aldrich; H6501) was prepared and KCl was added to a concentration of 60 mM to prevent hypokalemia. The β-3-OHB solution was infused continuously throughout two hours through the venous catheter with an infusion rate of 0.35 g/kg/h (approximately 14 g/h). This dose was empirically titrated to achieve plasma ketone concentrations comparable to those used in human studies17,24. To avoid any cofounding by endogeneous glucose production a hyperinsulinemic euglycemic glucose clamp (“HEC”)25 was used to maintain blood glucose around 5 mmol/L throughout the experiment.

The pigs underwent a baseline scan with hyperpolarized [2-13C]pyruvate (t = 0 min) and subsequently the infusion of β-3-OHB and the HEC was initiated. β-OHB was administered for two hours followed by a second scan (t = 120 min). The ketone infusion was then stopped while the HEC was continued for one hour followed by a third scan (t = 180 min). The study experimental timeline is depicted in Fig. 1..

A cartoon of this study’s experimental timeline.

The circulating glucose and β-OHB levels were monitored every 10 min using arterial blood with a portable electronic device (FreeStyle Precision Neo, Abbott Laboratories A/S, Copenhagen, Denmark). Arterial blood gas analysis was done every 30 min and whole blood samples were taken at baseline and every hour until the end of the experiment. After this point, pigs were euthanized via anesthetic overdose (pentobarbital potassium salt; Euthanimal, Alfasan Nederland BV, The Netherlands; equivalent to 365 mg/ml pentobarbitone solution, 100 mg/ml/kg dose) following the end of the experiment: i.e. all experiments were conducted under ultimately terminal anesthesia.

MR was performed on a clinical 3T GE Discovery 750 MR scanner (GE Healthcare, Milwaukee, WI, USA) with hyperpolarized [2-13C]pyruvate followed by a slice-selective, time-resolved spectroscopic acquisition as previously described (for additional information, please refer to supplemental information). Proton cardiac CINE images were additionally acquired, and the 12 cm spectroscopic slab was positioned (oblique if required) to encompass the whole heart. We note that the long axis of the porcine heart is typically perpendicular to the spine26, meaning that slice-selective spectra are likely well localized to the heart. We elected to perform spectroscopy rather than an imaging experiment in the absence of any expected localized alterations in cardiac metabolism.

Analysis of the CINE DICOM images was performed in the freely available software Segment (version 3.1 R8215, Lund: Medviso). ROIs were manually drawn in every timeframe of each scan to examine the myocardium of the left and right ventricle. ROIs were drawn and analyzed by the same investigator.

The multi-coil [2-13C]pyruvate spectroscopic data were reconstructed in MATLAB (2021b; The MathWorks Inc, Natick, Massachusetts) and peaks were quantified in OXSA27 using the AMARES algorithm28; after temporal summation of data for 90 s following injection; small metabolite peaks were not routinely quantifiable with adequate SNR to perform a kinetic modelling approach. Prior knowledge used in this quantification is provided in the supplemental information. The subsequently reported quantities are “AUC to pyruvate ratios”, that is, the temporal integral of the quantified spectral peaks (the “area-under-the-curve”) divided by the corresponding value for the visible [2-13C]pyruvate peak, to correct for shot-to-shot polarization differences.

Statistical analyses were performed in GraphPad Prism version 9.1 for Windows, (GraphPad Software, San Diego, California). One-way repeated-measures ANOVA was used with the Geisser-Greenhouse correction for sphericity followed by post-hoc t-tests multiplicity adjusted via the Holm-Sidak method, and results plotted as mean ± standard deviation. A p-value < 0.05 was considered statistically significant.

Hemodynamic, cardiac MRI and metabolic statistics are summarized in detail in Table 1. Figure 2 illustrates example spectroscopic data from one pig and highlights the main changes in [2-13C]pyruvate metabolism—a shift towards lactate production following pyruvate infusion during ketone and HEC infusion (“ketone + HEC” infusion).

Schematic summarizing the experimental process and a cartoon of the relevant metabolic pathways. (a) Here three timepoints for [2-13C]pyruvate are presented and the clear spectroscopic increase in labelled lactate during ketone + HEC is indicated. (b) The summed normalized spectra were of high SNR and permitted the quantification of multiple downstream metabolites, again with apparent changes in lactate production. The main peaks are the [2-13C]pyruvate tracer and its downstream metabolic products, lactate, acetyl-carnitine and glutamate are indicated by red and green arrows. (c) After temporal summation, spectra were quantified by AMARES utilizing [2-13C]pyruvate as a frequency reference; we elected to quantify summed spectra because the comparatively low SNR of metabolites in each timepoint hindered accurate quantification. (d) To highlight the most important findings in the study lactate (red box), acetyl-carnitine (green box) and glutamate (green box) are shown in the context of the TCA cycle; injected pyruvate can be seen to enter the TCA cycle if PDH flux is significant, leading to the quantification of downstream metabolites. If PDH flux is inhibited but pyruvate uptake through MCTs unchanged then greater label exchange into lactate would be predicted to occur.

Hemodynamic measurements are summarized in Fig. 3. Heart rate and cardiac output showed an increase during ketone infusion when comparing baseline to ketone + HEC, as well as comparing baseline to HEC. Heart rate increased from 51 to 85 bpm (p = 0.0004) and cardiac output from 1.9 to 3.8 L/min (p = 0.00108) after 2 h of ketone infusion; both additionally increased significantly between baseline and HEC (p = 0.0001; p = 0.00108 respectively). There were no significant differences in HR or CO between ketone + HEC and HEC alone (p = 0.275; p = 0.09). No significant differences were observed between the groups with respect to the observed left ventricular ejection fraction, blood pressure, or ventricular volumes.

Hemodynamics measurements during the experiments. Each hemodynamic parameter is plotted against baseline at 0 h, after 2 h of ketone infusion during HEC, and at 3 h, i.e.1 h after the cessation of ketone infusion. From top left: Heart rate, Systolic blood pressure (BP), Diastolic BP, Stroke volume, Ejection fraction (EF), Cardiac Output, end-diastolic and end-systolic volumes (EDV/ESV) of the left ventricle (LV). All graphs depict n = 5 measurements, each from a different pig. One-way repeated-measures ANOVA was used with the Geisser-Greenhouse correction for sphericity followed by post-hoc t-tests, with multiplicity adjusted via the Holm-Sidak method: ns denotes p ≥ 0.05; *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 10−4.

As shown in Fig. 4, many dynamic changes in hyperpolarized [2-13C]pyruvate metabolism were observed during the ketone infusion with HEC. Metabolic data from hyperpolarized [2‐13C]pyruvate MRI showed a decrease of apparent glutamate and acetyl-carnitine production, and an increase of apparent 13C lactate label exchange during the ketone infusion A significant decrease in 13C-Glutamate was observed between the baseline scan and the ketone + HEC (p = 0.033), and a significant increase in glutamate between the ketone + HEC and HEC alone (p = 0.033). Also, a significant decrease in glutamate from HEC to ketone + HEC was observed (p < 0.0001). 13C-acetylcarnitine showed an increase of acetylcarnitine from baseline to ketone + HEC (p = 0.0478). A significant decrease was also observed between the states of ketone + HEC to HEC alone (p = 0.0081). Accordingly, the derived ratio of 13C glutamate to 13C acetylcarnitine was significantly decreased between baseline and HEC (p = 0.014) and increased between ketone + HEC and HEC alone (p < 0.0009), but not significantly different between baseline and the state of ketone + HEC (p = 0.05001).

Graphs showing the presence of different metabolites in the myocardium at baseline, after 2 h ketone + HEC, and after 1 h on HEC alone. From left: Glutamate, acetyl-carnitine, lactate, alanine, and the glutamate to acetylcarnitine ratio. AMARES was used for MRS peak quantification following temporal integration for the presented ratios of metabolites compared to injected carbon (excluding the glutamate to acetylcarnitine ratio). All graphs depict n = 5 measurements, each from a different pig. One-way repeated-measures ANOVA was used with the Geisser-Greenhouse correction for sphericity followed by post-hoc t-tests, with multiplicity adjusted via the Holm-Sidak method: ns denotes p ≥ 0.05; *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 10−4.

13C-lactate was significantly increased from baseline to ketone + HEC (p = 0.0196), and decreased when comparing ketone + HEC to HEC alone (p = 0.0219). 13C-alanine was highly variable and displayed no significant differences between the conditions.

Data from blood gas analysis are depicted in Fig. 5. As expected on a HEC, glucose concentration remained comparatively stable, with variations from 4.1 mmol/L to 8.7 mmol/L, and always remaining within the normal range for blood glucose for pigs under anaesthesia29. Ketone body concentration in the blood increased within the first 30 min and was maintained at a high concentration for the whole period of ketone infusion with a clear washout period: after the termination of the ketone infusion at 120 min, ketone concentration decreased rapidly with a detectable plasma concentration almost at 0 mmol/L one hour after the termination of the infusion. Lactate blood concentration was variable, without a significant simple linear trend either throughout the entire experiment (p = 0.11, linear regression) or during the period of the ketone infusion alone (p = 0.76). Potassium concentration followed a linear trend throughout the experiment with a small but statistically significant decrease throughout the duration of the experiment (simple linear regression; R2 = 0.77; p < 0.0001; mean rate of decrease 0.0059 mM [K+]/min); calcium was unchanged (p = 0.099); and a small but statistically significant increase in pH and sodium concentration was similarly observed (R2 = 0.65, 0.38; p < 0.0001, p = 0.0005); mean rate of increase 0.00059 pH units / min; 0.027 mM [Na+]/min; n = 4 for sodium owing to a technical failure). There was no significant linear trend in pCO2 (p = 0.87).

Arterially sampled blood data plotted against time, showing the effects on a number of biomarkers after ketone infusion and HEC: (a) Glucose, (b) Ketone, (c) Lactate, (d) pH, (e) Potassium, (f) Calcium, (g) Sodium and h) the partial pressure of carbon dioxide. The dotted lines indicate, from left to right, the start of the ketone + HEC, the end of the ketone infusion, and the end of the experiment. All graphs except that for sodium depict n = 5 measurements, each from a different pig; n = 4 for the Na+ measurement owing to a technical error. As detailed in the text, a significant simple linear trend is present showing an increase in pH and Na+ over the experiment (R2 = 0.65, 0.38; p < 0.0001, p = 0.0005; mean rate of increase 0.00059 pH units / min; 0.027 mM [Na+]/min respectively), and a decrease in K+ (R2 = 0.77; p < 0.0001; mean rate of decrease 0.0059 mM [K+]/min). The mean value across all animals ± 1 standard deviation is shown in red.

This study is the first using hyperpolarized 13C magnetic resonance spectroscopy to show in real time that the infusion of the ketone body β-3-OHB leads to profound cardiac metabolic shifts. In it, we have shown that ketone administration significantly altered both the hemodynamics and metabolism of the porcine heart. We observed an approximate doubling of cardiac output during the ketone infusion and HEC that persisted for at least an hour beyond the end of the ketone infusion.

This observation is consistent with clinical studies, and suggestive of either a decrease in systemic vascular resistance, increased myocardial contractility, or both. The combination of insulin and potassium chloride administered IV at these doses is known to have vasodilatory effects30,31. Furthermore, β-3-OHB is itself also a potent vasodilator that affects endothelial nitric oxide release, and is reported to exhibit vasodilatory effects in animal models32,33,34. Peripheral vascular resistance therefore is expected to decrease under these experimental conditions, and the baroreflex would maintain blood pressure by consequently raising heart rate and thus cardiac output. Both changes were observed. This physiological homeostatic response is only possible, however, if cardiac metabolism is flexible enough to permit the heart to meet its new energetic demand. The finding of an increase in cardiac output following two hours of ketone infusion in pigs confirms the previously reported result from ketone infusion in humans utilizing the same preparation, in which the HEC alone does not cause a corresponding increase in cardiac output in patients with heart failure with reduced ejection fraction (HFrEF)17,18.

Following the infusion of ketones, the downstream TCA-cycle metabolites glutamate and acetyl-carnitine were decreased and upregulated again one hour after the end of the infusion, whereas lactate levels in the myocardium were upregulated in response to ketone infusion and downregulated after its termination. In the situation where the HEC is present, reflecting the “mildly fed” state in which glucose and insulin are both available, pyruvate enters the cell through monocarboxylate transporters, and enters the TCA cycle as glucose is being oxidized to CO2. We therefore observe a basal level of acetyl-carnitine label incorporation, reflecting the role of carnitine as a buffer of CoA through carnitine O-acetyltransferase, and the net flux of the acetyl-CoA moiety into the TCA cycle, and corresponding glutamate production. These conditions also reflect a baseline level of hydraulic work. During the ketone body infusion with HEC, β-3-OHB enters the cell through monocarboxylate transporters and is converted into acetoacetate via β-hydroxybutyrate dehydrogenase (βHBDH), with a rate that is very dependent on the ratio of NAD+/NADH, with the more reduced state favoring β-OHB formation and the more oxidized state favoring acetoacetate. Under the conditions of GI infusion, this intracellular ratio likely shifts9, meaning that effectively if β-3-OHB is present it can be utilized, powering the TCA cycle. We observe this substrate selection through the concomitant increase in labelled lactate being produced from hyperpolarized [2-13C]pyruvate: acetyl-CoA moieties inhibit pyruvate dehydrogenase, shunt the incoming “glycolytic” pyruvate flux towards lactate, resulting in a lower label incorporation fraction to acetyl-carnitine and glutamate as less pyruvate enters the cell. These shifts in substrate selection can occur independent of other forms of regulation in the TCA cycle as βHBDH has far fewer known points of allosteric or other forms of regulatory control12, and ketone bodies could be effectively replenishing any extant anaplerotic flux towards the TCA cycle that is otherwise required. However, perfused rat heart experiments on acetoacetate alone show that some degree of pyruvate oxidation is required to prevent a decline of contractile function that may arise due to the lack of suitable anaplerotic pathways35. Despite the shift towards ketone body utilization during the infusion, the existing glucose metabolism most likely remains thermodynamically favorable (i.e., the perfused rat heart has a reduction in the magnitude of Gibbs free energy, ∆G, from glucose oxidation but it remains negative and thus favored9). Therefore, the overall effect is a matching of the increased fuel supply from both glucose and ketone bodies to meet the increased energy demands imposed by the elevated cardiac output resulting from vasodilation. Excess substrates produced by this experiment would also likely feed into the Cahill (glucose/alanine) and Cori (glucose/lactate) cycles, where hepatic metabolism would be altered accordingly.

After the end of the ketone infusion, a second dose of [2-13C]pyruvate showed a normalization of label exchange into lactate and acetyl-carnitine and an increase in glutamate compared to pre-infusion. We propose that this is consistent with the restoration of “normal” fed metabolism in the situation in which cardiac output is increased via decrease in vascular resistance, and oxidative metabolism is able to meet this demand. Further work with “metabolically inflexible” hearts (i.e. in HF) assessed by hyperpolarized MRI may well be able to elucidate the cardiac metabolic derangement associated with different pathophysiology such as heart failure.

Other interventional studies have also shown that cardiac metabolism increases in correlation to increased cardiac output and cardiac work, but these studies show specifically an overall increase in metabolism rather than a change of metabolic pathways within the myocardium as seen in this study, consistent with the hypothesis of some degree of vasodilation but without the specificity to probe particular alterations in fluxes through pathways36,37.

We aimed to control all relevant biochemical parameters through blood monitoring and dietary control. Hence, all pigs were fed the same diet prior to the experiment. Blood glucose concentration was maintained at the same level by using the HEC, to eliminate the effects of glucose on cardiac metabolism and fully evaluate the effect of ketone bodies on cardiac metabolism and function. The whole-blood concentration of ketones increased progressively as we infused ketone solution and dropped rapidly to almost undetectable levels within an hour after termination of the infusion. However, the lactate concentration in (whole) blood increased within the first hour of infusion and then remained stable throughout the study. It is well described in other studies using the HEC that insulin stimulates glycolysis, which in turn favors the conversion of pyruvate to lactate23, increasing the lactate concentration. This can explain why lactate blood levels rise as the experiment begins and remains at the same level even after ketone infusion stopped. Yet our study shows an interesting change in the metabolic pathways specific to the heart as the rate of 13C label incorporation into lactate as measured by MRI follows the infusion of ketones; an increase in lactate while infusing ketones and a decrease after the termination of the infusion even though the HEC is retained. This is consistent with an organ-specific metabolic shuttle occurring in the myocardium itself, and that PDH flux (and thus activity38) is altered by ketones. One further subtlety of hyperpolarized [2-13C]pyruvate experiments is the direct interpretation of the label exchange that they observe: since [2-13C]pyruvate must pass through the highly regulated enzyme complex of pyruvate dehydrogenase (PDH) to enable the 13C label to enter the TCA cycle, the production of downstream metabolites is only visible if PDH flux occurs. Therefore, our key findings likely reflect the suppression of glucose oxidation (and hence PDH flux) leading to lower 13C label entry despite augmented TCA cycle metabolism from additional ketone fuels, which we do not directly measure. The use of hyperpolarized, labelled, β-3-OHB or acetoacetate39,40,41 may show apparent rates of oxidation of these substrates, but for the reasons given above, we argue that there are fewer points of regulation on their metabolism and as probes they may more accurately reflect particular redox couples and probe availability rather than the metabolic fate of ketones already present directly. In this situation, the realtime readout of the control of PDH provided by the use of [2-13C]pyruvate is key to understanding substrate switching that occurs when ketones are made available.

In this work, cardiac output and heart rate remained increased compared to baseline for at least an hour after the termination of the ketone infusion. This may indicate that the biochemical and hemodynamic effects due to ketone infusion remain for a longer time than that of active infusion; the effects also remain in a state where the ketone concentration in the blood is negligible. This study is obviously limited in its timeframe and does not evaluate the effects of ketone bodies beyond 3 h, although it is known that β-3-OHB itself has signaling effects that may differ between the L- and D-isomers42. Further studies are therefore needed to determine the long-term effect; it may be possible to separately administer L- or D-isomers and identify a distinction between metabolic and direct signaling effects; another alternative may be to administer acetoacetate (AcAc) directly, although AcAc is chemically unstable and typically needs to be regularly synthesized through a multi-step synthetic pathway prior to use in preclinical studies39,43. Here, we also used a slightly more concentrated ketone solution in this experiment than that of other studies performed in humans17,18,34. This change was determined by an empirical dose titration to reach the same blood ketone concentration as used previously in humans and similar to other studies in pigs44. An important limitation of this work is that we did not administer a volume-matched placebo infusion following termination of the β-3-OHB itself—this may have caused a slight relative decrease in cardiac preload at the third timepoint and potentially affected haemodynamic measurements such as HR, SV, CO and thus energetics. However, no changes in EDV were observed throughout the study, and there was no significant change in HR between the periods of ketone + HEC and HEC alone, indicating that this effect is likely minimal, at least during the timescale of these experiments. Another confounding factor we could not control for is that propofol is highly lipophilic and administered as a soybean oil emulsion; we therefore co-administered 50 mg/ml medium chain triglycerides and 50 mg/ml long chain triglycerides for a total lipid infusion of 1.5 g/hour. This is an order of magnitude lower than the rate of infusion of other substrates (i.e. approximately 14 g/hour ketone) and likely lower than basal rates of endogenous triglyceride secretion in either fasted or fed states (reported as ~ 100 mg/kg/hour in monkeys)45. Another unavoidable confounding effect is that of anaesthesia itself; previous hyperpolarized 13C reports note that anaesthetic agent and dose influences the metabolism of pyruvate in both heart and brain and emphasise the importance of utilizing a standardized protocols46,47,48. Propofol is one of the world’s most widely used anaesthetics, including in the context of cardiac surgery49, and our protocol here is consistent with recommended guidelines50, but its direct metabolic effects are less well known.

The cardiovascular effects of ketones have already shown great promise in heart failure patients, but they might also have a usage in other aspects of heart disease and treatment. The beneficial effects of ketones on cardiac function could be transferred to prophylactic treatment of diabetes patients or into heart transplantation, where an increasing demand and low supply of donor hearts have led to a shortage of donor organs for transplantation51,52. Thus, it is important to determine the physiological effects of ketones, in large-animal disease models. This study provides new insights on how the cardiac metabolism reacts to exogenous ketone supplementation, and brings us one step closer to fully understand ketones and their interventional potential.

We have demonstrated that infusion with the ketone body, β-3-OHB causes profound changes in energetically important myocardial metabolic intermediates, through non-invasive assessment using hyperpolarized [2-13C]pyruvate magnetic resonance spectroscopy. However, more studies are warranted to determine the longer-term effects of ketone bodies on cardiac metabolism and function, including in large animal models of disease.

The detailed datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request; detailed summary statistics are provided in Table 1.

Møller, N. Ketone body, 3-hydroxybutyrate: minor metabolite - major medical manifestations. J. Clin. Endocrinol. Metab. 105, dgaa370 (2020).

Article PubMed Google Scholar

Janardhan, A., Chen, J. & Crawford, P. A. Altered systemic ketone body metabolism in advanced heart failure. Tex. Heart Inst. J. 38, 533–538 (2011).

PubMed PubMed Central Google Scholar

Abdurrachim, D. et al. Empagliflozin reduces myocardial ketone utilization while preserving glucose utilization in diabetic hypertensive heart disease: a hyperpolarized 13 C magnetic resonance spectroscopy study. Diabetes Obes. Metab. 21, 357–365 (2019).

Article CAS PubMed Google Scholar

Kappel, B. A. et al. Effect of empagliflozin on the metabolic signature of patients with type 2 diabetes mellitus and cardiovascular disease. Circulation 136, 969–972 (2017).

Article CAS PubMed MATH Google Scholar

Mudaliar, S., Alloju, S. & Henry, R. R. Can a shift in fuel energetics explain the beneficial cardiorenal outcomes in the EMPA-REG OUTCOME study? A unifying hypothesis. Diabetes Care. 39, 1115–1122 (2016).

Article CAS PubMed Google Scholar

Nishimura, R. et al. Effect of empagliflozin on free fatty acids and ketone bodies in Japanese patients with type 2 diabetes mellitus: a randomized controlled trial. Adv. Ther. 36, 2769–2782 (2019).

Article CAS PubMed MATH Google Scholar

Santos -Gallego Carlos, G. et al. Empagliflozin induces a myocardial metabolic shift from glucose consumption to ketone metabolism that mitigates adverse cardiac remodeling and improves myocardial contractility. J. Am. Coll. Cardiol. 71, A674–A674 (2018).

Article MATH Google Scholar

Seefeldt, J. M. et al. Cardioprotective effects of empagliflozin after ischemia and reperfusion in rats. Sci. Rep. 11, 9544 (2021).

Article ADS CAS PubMed PubMed Central MATH Google Scholar

Sato, K. et al. Insulin, ketone bodies, and mitochondrial energy transduction. FASEB J. 9, 651–658 (1995).

Article CAS PubMed MATH Google Scholar

Cox, P. J. J. et al. Nutritional ketosis alters fuel preference and thereby endurance performance in athletes. Cell Metab. 24, 256–268 (2016).

Article CAS PubMed Google Scholar

Carneiro, L. et al. Evidence for hypothalamic ketone body sensing: impact on food intake and peripheral metabolic responses in mice. Am. J. Physiol. Endocrinol. Metab. 310, E103–E115 (2016).

Article PubMed MATH Google Scholar

Deja, S. et al. In vivo estimation of ketogenesis using metabolic flux analysis—technical aspects and model interpretation. Metabolites 11, 279 (2021).

Article CAS PubMed PubMed Central MATH Google Scholar

Neubauer, S. The failing heart—an engine out of fuel. N. Engl. J. Med. 356, 1140–1151 (2007).

Article PubMed MATH Google Scholar

Schroeder, M. A. et al. Hyperpolarized (13)C magnetic resonance reveals early- and late-onset changes to in vivo pyruvate metabolism in the failing heart. Eur. J. Heart Fail. 15, 130–140 (2013).

Article CAS PubMed Google Scholar

Doenst, T., Nguyen, T. D. & Abel, E. D. Cardiac metabolism in heart failure: implications beyond ATP production. Circ. Res. 113, 709–724 (2013).

Article CAS PubMed PubMed Central Google Scholar

Burrage, M. K., Lewis, A. J. & Miller, J. J. J. Functional and metabolic imaging in heart failure with preserved ejection fraction: promises, challenges, and clinical utility. Cardiovasc. Drugs Ther. https://doi.org/10.1007/s10557-022-07355-7 (2022).

Article PubMed PubMed Central MATH Google Scholar

Roni Nielsen, N. et al. Cardiovascular effects of treatment with the ketone body 3-hydroxybutyrate in chronic heart failure patients. Circulation 139, 2129–2141 (2019).

Article PubMed PubMed Central MATH Google Scholar

Gormsen, L. C. et al. Ketone body infusion with 3-hydroxybutyrate reduces myocardial glucose uptake and increases blood flow in humans: a positron emission tomography study. J. Am. Heart Assoc. https://doi.org/10.1161/jaha.116.005066 (2017).

Article PubMed PubMed Central MATH Google Scholar

Berg-Hansen, K. et al. Beneficial effects of ketone ester in patients with cardiogenic shock: a randomized, controlled, double-blind trial. JACC Heart Fail. 11, 1337–1347 (2023).

Article CAS PubMed MATH Google Scholar

Berg-Hansen, K. et al. Cardiovascular effects of oral ketone ester treatment in patients with heart failure with reduced ejection fraction: a randomized, controlled, double-blind trial. Circulation 149, 1474–1489 (2024).

Article CAS PubMed PubMed Central MATH Google Scholar

Tougaard, R. S. et al. Hyperpolarized [1-13C]pyruvate MRI can image the metabolic shift in cardiac metabolism between the fasted and fed state in a porcine model. Magn. Reson. Med. 81, 2655–2665 (2019).

Article CAS PubMed MATH Google Scholar

Josan, S. et al. In vivo investigation of cardiac metabolism in the rat using MRS of hyperpolarized [1-13C] and [2-13C] pyruvate. NMR Biomed. 26, 1680–1687 (2013).

Article CAS PubMed MATH Google Scholar

Hansen, E. S. S. et al. Imaging porcine cardiac substrate selection modulations by glucose, insulin and potassium intervention: a hyperpolarized [1-13 C]pyruvate study. NMR Biomed. 30, e3702 (2017).

Article Google Scholar

Stubbs, B. J. et al. On the metabolism of exogenous ketones in humans. Front. Physiol. 8, 848 (2017).

Article PubMed PubMed Central MATH Google Scholar

Bengtsen, M. B. et al. Hyperpolarized [1-13 C]pyruvate combined with the hyperinsulinaemic euglycaemic and hypoglycaemic clamp technique in skeletal muscle in a large animal model. Exp. Physiol. 106, 2412–2422 (2021).

Article CAS PubMed MATH Google Scholar

Crick, S. J. et al. Anatomy of the pig heart: comparisons with normal human cardiac structure. J. Anat. 193, 105–119 (1998).

Article PubMed PubMed Central MATH Google Scholar

Purvis, L. A. B. et al. An open-source magnetic resonance spectroscopy analysis toolbox in MATLAB. PLoS ONE 12. https://doi.org/10.1371/journal.pone.0185356 (2017).

Vanhamme, L., van den Boogaart, A. & Van Huffel, S. Improved method for accurate and efficient quantification of MRS data with use of prior knowledge. J. Magn. Reson. (San Diego Calif: 1997) 129, 35–43 (1997).

Article ADS CAS MATH Google Scholar

Manell, E., Jensen-Waern, M. & Hedenqvist, P. Anaesthesia and changes in parameters that reflect glucose metabolism in pigs—a pilot study. Lab. Anim. 51, 509–517 (2017).

Article CAS PubMed Google Scholar

Wambach, G. K. & Liu, D. Insulin attenuates vasoconstriction by noradrenaline, serotonin and potassium chloride in rat mesenteric arterioles. Clin. Exp. Hypertens. Part. Theory Pract. 14, 733–740 (1992).

Article CAS Google Scholar

Yki-Järvinen, H. & Utriainen, T. Insulin-induced vasodilatation: physiology or pharmacology? Diabetologia 41, 369–379 (1998).

Article PubMed Google Scholar

McCarthy, C. G. et al. Ketone body β-hydroxybutyrate is an autophagy-dependent vasodilator. JCI Insight 6, e149037.

Homilius, C. et al. Ketone body 3-hydroxybutyrate elevates cardiac output through peripheral vasorelaxation and enhanced cardiac contractility. Basic Res. Cardiol. 118, 37 (2023).

Article CAS PubMed PubMed Central Google Scholar

Nielsen, R. et al. Hemodynamic effects of ketone bodies in patients with pulmonary hypertension. J. Am. Heart Assoc. https://doi.org/10.1161/JAHA.122.028232 (2023).

Article PubMed PubMed Central MATH Google Scholar

Russell, R. R. & Taegtmeyer, H. Pyruvate carboxylation prevents the decline in contractile function of rat hearts oxidizing acetoacetate. Am. J. Physiol. -Heart Circ. Physiol. 261, H1756–H1762 (1991).

Article CAS Google Scholar

Bøgh, N. et al. Increasing carbohydrate oxidation improves contractile reserves and prevents hypertrophy in porcine right heart failure. Sci. Rep. 10, 8158 (2020).

Article ADS PubMed PubMed Central MATH Google Scholar

Joergensen, S. H. et al. Detection of increased pyruvate dehydrogenase flux in the human heart during adenosine stress test using hyperpolarized [1-13 C]pyruvate cardiovascular magnetic resonance imaging. J. Cardiovasc. Magn. Reson. 24, 34 (2022).

Article PubMed PubMed Central Google Scholar

Atherton, H. J. et al. Validation of the in vivo assessment of pyruvate dehydrogenase activity using hyperpolarised 13 C MRS. NMR Biomed. 24, 201–208 (2011).

Article CAS PubMed MATH Google Scholar

Miller, J. J. et al. Hyperpolarized ketone body metabolism in the rat heart. NMR Biomed. https://doi.org/10.1002/nbm.3912 (2018).

Article PubMed PubMed Central MATH Google Scholar

Chen, W. et al. Metabolism of hyperpolarized 13 C-acetoacetate/ß-hydroxybutyrate reveals mitochondrial redox state in perfused rat hearts. Proc. Int. Soc. Magn. Reson. Med. 24, 0672 (2016).

Abdurrachim, D. et al. A new hyperpolarized 13 C ketone body probe reveals an increase in acetoacetate utilization in the diabetic rat heart. Sci. Rep. https://doi.org/10.1038/s41598-019-39378-w (2019).

Article PubMed PubMed Central Google Scholar

Newman, J. C. & Verdin, E. β-Hydroxybutyrate: a signaling metabolite. Annu. Rev. Nutr. 37, 51–76 (2017).

Article CAS PubMed PubMed Central MATH Google Scholar

Hall, L. M. Preparation of crystalline lithium acetoacetate. Anal. Biochem. 3, 75–80 (1962).

Article CAS PubMed MATH Google Scholar

Gopalasingam, N. et al. Enantiomer-specific cardiovascular effects of the ketone body 3‐hydroxybutyrate. J. Am. Heart Assoc. 13, e033628 (2024).

Article CAS PubMed PubMed Central Google Scholar

Fiser, R. H. Jr et al. Triglyceride secretion rates: use of Triton WR 1339 in the rhesus monkey. J. Nutr. 104, 223–226 (1974).

Article CAS PubMed MATH Google Scholar

Healicon, R. et al. Assessing the effect of anesthetic gas mixtures on hyperpolarized 13 C pyruvate metabolism in the rat brain. Magn. Reson. Med. 88, 1324–1332 (2022).

Article CAS PubMed PubMed Central MATH Google Scholar

Steinhauser, J. et al. Assessing the influence of isoflurane anesthesia on cardiac metabolism using hyperpolarized [1-13 C]pyruvate. NMR Biomed. 31, e3856 (2018).

Article MATH Google Scholar

Hyppönen, V. et al. Metabolism of hyperpolarised [1–13 C]pyruvate in awake and anaesthetised rat brains. NMR Biomed. 35, e4635 (2022).

Article PubMed Google Scholar

Pasin, L. et al. Propofol and survival: a meta-analysis of randomized clinical trials. Acta Anaesthesiol. Scand. 59, 17–24 (2015).

Article CAS PubMed MATH Google Scholar

Costea, R., Ene, I. & Pavel, R. Pig sedation and anesthesia for medical research. Animals (Basel) 13, 3807 (2023).

Article PubMed MATH Google Scholar

Levitt, M. Could the organ shortage ever be met? Life Sci. Soc. Policy 11, 6 (2015).

Article PubMed PubMed Central MATH Google Scholar

Seefeldt, J. M. et al. Effects of ketone body 3-hydroxybutyrate on cardiac and mitochondrial function during donation after circulatory death heart transplantation. Sci. Rep. 14, 757 (2024).

Article ADS CAS PubMed PubMed Central Google Scholar

Download references

This work was supported by the Lundbeck Fonden (all authors through Aarhus Universitet) and the Novo Foundation, grant number NNF21OC0068683 to JM.

Christoffer Laustsen and Jack J. Miller contributed equally.

The MR Research Centre, Aarhus University, Aarhus, Denmark

Sabrina Kahina Bech, Esben Søvsø Szocska Hansen, Christoffer Laustsen & Jack J. Miller

Department of Cardiology, Aarhus University Hospital, Aarhus, Denmark

Bent Roni Nielsen & Henrik Wiggers

Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Aarhus, Denmark

Mads Bisgaard Bengtsen

You can also search for this author in PubMed Google Scholar

SKB and ESSH conducted experiments with input and assistance from BRN, HW, MBB and CL. SKB and JJM wrote the main manuscript text and prepared figures, and all authors assisted with interpretation, reviewing the manuscript and drafting.

Correspondence to Jack J. Miller.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Below is the link to the electronic supplementary material.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

Bech, S.K., Hansen, E.S.S., Nielsen, B.R. et al. Myocardial metabolic flexibility following ketone infusion demonstrated by hyperpolarized [2-13C]pyruvate MRS in pigs. Sci Rep 15, 5849 (2025). https://doi.org/10.1038/s41598-025-90215-9

Download citation

Received: 22 November 2024

Accepted: 11 February 2025

Published: 18 February 2025

DOI: https://doi.org/10.1038/s41598-025-90215-9

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative