Article Text

other Versions

Non-invasive imaging
Metabolic imaging of the human heart: clinical application of magnetic resonance spectroscopy
  1. Maurice B Bizino,
  2. Sebastiaan Hammer,
  3. Hildo J Lamb
  1. Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands
  1. Correspondence to Maurice B Bizino, Department of Radiology, Leiden University Medical Centre, Albinusdreef 2, Leiden 2333 ZA, The Netherlands; m.b.bizino{at}

Statistics from

Cardiovascular MRI has earned its place in the field of clinical cardiac imaging. Regularly used techniques include anatomical imaging, functional imaging, perfusion, and delayed enhancement (DE). Cardiac magnetic resonance spectroscopy (MRS) uses the same hardware, measuring the abundance of metabolites in the myocardium in vivo non-invasively without the use of radiation or external tracers. Its main application is currently scientific to gain insight into metabolic changes in cardiac pathologies.

The heart is a metabolically active organ using on average 6 kg of adenosine triphosphate (ATP) each day. As energy is crucial for both systole and diastole, derangements in energy metabolism may be the first step in failure of the heart.1 By using the gyromagnetic properties of 1H, 31P, 13C, and 23Na, MRS is a powerful tool to relate energy metabolism to (dys)function of the heart.

The aim of this article is to provide an overview of the current use, opportunities and limitations of MRS in relation to common cardiac diseases: ischaemic heart disease, heart failure, inherited cardiomyopathy, the metabolic syndrome, valvular heart disease, and heart transplantation.

Cardiac energy metabolism

A simplified schematic representation of cardiac energy metabolism and opportunities to assess components of metabolism with MRS is depicted in figure 1. On average, the heart cycles 10 tons of blood each day in 100 000 heart beats. To meet the enormous ATP requirement, cardiomyocytes fuel themselves with free fatty acids (FFA) and glucose as the primary source of chemical energy. FFA and glucose contribute to ATP synthesis in terms of supply of chemical energy in a ratio of 3:1 in normal situations. Derangements in substrate utilisation are associated with a wide variety of diseases which will be discussed below. The uptake of FFA by the fatty acid transporter is an energy consuming process. Fatty acids enter the mitochondrion where β-oxidation takes place after which the intermediate acetyl-coenzyme A (CoA) enters the Krebs cycle. The uptake of glucose by the glucose transporter type 4 (GLUT4) is insulin dependent. Glucose is converted to pyruvate in the cytoplasm by glycolysis. Pyruvate enters the Krebs cycle in the mitochondrion (figure 1).

Figure 1

Magnetic resonance spectroscopy (MRS) assessment of cardiac energy metabolism. Upper figure depicts a schematic diagram of energy metabolism of the cardiomyocyte. In the mitochondrion (blue) ATP is generated from mainly glucose and FFAs. ATP and regulation by calcium flux are needed for contraction and relaxation of the myofibrils. The coloured boxes (yellow, green and red) indicate opportunities for measurement with MRS, respectively 1H, 31P, and 13C. Glucose and FFAs are the primary sources of chemical energy. Glucose uptake by the GLUT4 is insulin dependent; FFAs are transported into the cell via the FFA transporter. Glucose is subject to glycolysis and converted to pyruvate which enters the mitochondrion or is converted to lactate. FFAs are either stored as triglycerides or enter the mitochondrion. Both pyruvate and FFA (after β-oxidation) give rise to acetyl-CoA. Acetyl-CoA enters the Krebs cycle. In the mitochondrion the process of oxidative phosphorylation generates ATP. Outside the mitochondrion energy is transferred to PCr which releases its energy in the myofibrils giving rise to ATP and Cr. Acetyl-CoA, acetyl-coenzyme A; ADP, adenosine diphosphate; ATP, adenosine triphosphate; B-ox, β-oxidation; Ca2+, ionised calcium; Cr, creatine; FFA, free fatty acid; GLUT4, glucose transporter type 4; PCr, phosphocreatine; TG, triglyceride content.

The process of oxidative phosphorylation is the basis for production of energy by the mitochondrial respiratory chain. This yields energy needed for the generation of ATP from adenosine diphosphate (ADP). Transfer of chemical energy (ATP) from mitochondrion to myofibril is provided by the creatine kinase energy shuttle. In the mitochondrion, creatine kinase catalyses conversion of ATP and creatine to phosphocreatine (PCr) and ADP (figure 1). PCr rapidly diffuses to the myofibrils. Once there, creatine kinase catalyses the reaction back to ATP and free creatine. Creatine then diffuses back to the mitochondrion. Two thirds of the total creatine pool is phosphorylated to PCr; the other part remains in the cell as free creatine. The creatine kinase system acts as an important energy buffer, providing the heart with energy when the demand outweighs the supply. Such a situation results in a decreased PCr concentration and increased ADP concentration, whereas ATP concentrations are held constant. Overconsumption of PCr and subsequent increase in ADP lead to inhibition of many intracellular enzymes.1 Furthermore the process of energy production is regulated by calcium flux. For example, calcium flux from the sarcoplasmic reticulum to the mitochondrion can directly activate oxidative metabolism.w1 Both diastolic and systolic heart function are dependent upon energy synthesis and utilisation. This is reflected by the fact that calcium handling is an important determinant of ATP-consuming relaxation of cardiomyocytes as well as systolic function of the heart.w2

Metabolic assessment of the myocardium with 1H, 31P, 13C and 23Na—MRS

MRS of the heart has been performed using 1.5 and 3.0 tesla (T) MR systems, and recently even using ultra-high 7 T MR systems. Modern systems can be delivered with software to support MRS. Depending on the MR system (manufacturer, field strength) coils for 1H and 31P spectroscopy are commercially available. However, to initiate cardiac MRS, technical development and support is needed to optimise the procedure—that is, shimming techniques, gradients and pulse sequences. Furthermore, each MR system requires its own specific adjustments. For 13C and 23Na-MRS coils, pulse sequences and metabolic tracers are not routinely obtainable.

As the heart moves due to contraction and breathing, MRS is subject to motion artefacts. Therefore, ECG triggering is necessary to correct for motion of the heart throughout the cardiac cycle. Respiratory motion compensation has been performed with respirometer triggered acquisition, breath-hold scanning sequences, and by using navigator echoes.w3 w4 Figure 2 shows a navigator gated acquisition of an 1H spectrum of the heart. The scanning time for 1H-MRS varies from a single breath-hold to 20 min (excluding time required for 1H imaging for planning of the voxel). The most commonly used pulse sequences are stimulated echo acquisition mode (STEAM) and point resolved spectroscopy (PRESS). STEAM has the disadvantage of a lower signal-to-noise ratio (SNR) and increased susceptibility to motion, but enables acquisition with shorter echo time and hence better detection of fat. As SNR increases with field strength, 1H-MRS has recently been performed in one breath-hold at 3 T with STEAM.w3 Based on the required variability in the measurement, mode of respiratory compensation and scanning time, an appropriate method can be chosen. 31P-MRS using the method depicted in figure 3 requires about 10 min. For postprocessing the spectra jMRUI is the most commonly used programme. Table 1 shows an overview of the potential clinical applications of 1H, 31P, 13C, and 23Na-MRS and challenges of the technique.

Table 1

Summary of potential clinical applications of cardiac magnetic resonance spectroscopy

Figure 2

Navigator gated 1H magnetic resonance (MR) spectroscopy of the heart. Upper left panel shows preparation and monitoring of the respiratory navigator gating technique. A pencil beam is placed on the dome of the right hemidiaphragm on coronal (A) and transverse (B) MR images. White dots (C) represent the automatically traced position of the diaphragm during calibration. In (D) horizontal lines indicate the acceptance window; whenever the detected position of the diaphragm is within the predefined window, the spectroscopic measurement is stored. The upper right panel shows the acquisition of the spectrum. The surface coil is positioned just below the anatomical level of the mitral valve on sagittal (E) and transverse (F) MR images. By using a four chamber (G) and short axis (H) view the spectroscopic volume is placed in the interventricular septum. As water is by far the most abundant substance, an 1H spectrum as depicted in (I) can only be acquired after applying suppression of the water signal. Adapted from van der Meer et al.w4

Figure 3

31P-magnetic resonance spectroscopy of the human heart. Volume of interest is positioned in an optimal orientation relative to the left ventricle based on scout images in the transverse (A) and sagittal imaging plane (B). The volume (square box) is placed perpendicular to the chest wall to prevent contamination with skeletal muscle. By adjusting the level of the volume selection in the caudo-cranial direction, contamination of the sensitive volumes by diaphragm muscle and liver tissue can be prevented (B). The white square on the chest wall originates from a reference sample in the centre of the surface coil. With this technique a spectrum as shown in (C) can be acquired. Note that with this technique no regional differences can be detected, but an average of the signals in the left ventricle is acquired. 2,3 DPG, 2,3-diphosphoglycerate; PDE, phosphodiesters; PCr, phosphocreatine.

1H is the most suitable nucleus for MRS because it has the highest MR sensitivity. 1H-MRS can detect various metabolites in the myocardium that are associated with (dys)function (figure 1). Assessment of triglyceride (TG) content of the myocardium is of major importance in view of the epidemic proportions of cardiovascular disease related to obesity in modern society. Using 1H-MRS, myocardial TG content has been shown to be an independent predictor of diastolic dysfunction in both healthy menw5 and type 2 diabetes mellitus (T2DM) patients.2 Furthermore, creatine content can be determined with 1H-MRS. As part of the creatine kinase energy shuttle, creatine content reflects the energy status of the heart. In heart failure and non-viable myocardium creatine content is depleted.w6 w7 Other metabolites that can be detected are lactate, (de)oxymyoglobin, and carnitine.

31P-MRS has been used extensively to assess high energy phosphate metabolism which is crucial for maintaining normal function of the heart. 31P-MRS is therefore a useful tool to analyse the energy status of myocardium. The PCr peak and a combination of the three peaks of ATP are used to calculate the PCr/ATP ratio (figure 1). This ratio is the most commonly used parameter in 31P-MRS of the heart. In general there are two conditions associated with a decreased ratio: (1) ATP synthesis cannot meet the ATP requirement—that is, in ischaemia; and (2) decreased total creatine supply is diminished. Absolute measurements of PCr and ATP can also be assessed, albeit in a more complex way.1 Figure 3 shows an example of a technique to perform 31P-MRS of the heart and an example of a phosphorus spectrum.

13C-MRS is limited by the fact that it has too low MR sensitivity and an abundance in myocardial tissue to be used in vivo. However, with the exciting new technique of hyperpolarisation (dynamic nuclear polarisation technique) the signal can be increased >10 000 times. A hyperpolarised metabolic tracer is made by mixing the molecule with free radicals, placing it in a magnetic field, and then freezing it to a temperature close to 1 K.w8 In cardiometabolic research, 1-13C pyruvate, 2-13C pyruvate, and 13C-bicarbonate tracers have been used in animal studies, allowing analysis of pyruvate dehydrogenase flux and assessment of components of β-oxidation and the Krebs cycle (figure 1). Figure 4 shows that 13C-MRS could be used for detection of ischaemia and/or infarction. Although 1-13C pyruvate has been used in a research setting in one human study, technical issues regarding safety have to be addressed before the technique of hyperpolarisation gains regulatory approval for widespread use.

Figure 4

13C-magnetic resonance spectroscopy (MRS) combined with MRI of a pig heart. By combining proton MRI, perfusion and late gadolinium enhancement (LGE) with 13C MRS with a 1-13C pyruvate tracer, ischaemic and infarcted tissue can be characterised. These images were obtained in pig hearts. By superimposing the signals acquired in spectroscopy on proton MR images, maps can be made for different metabolites. Alanine reflects cytoplasmic cell metabolism whereas bicarbonate indicates mitochondrial activity. In the ischaemic model (15 min occlusion of left circumflex artery) only the bicarbonate changes, whereas in the infarction model (45 min of occlusion) the alanine signal drops and the bicarbonate signal almost disappears, spatially correlated to LGE area. Note that the spatial resolution is high enough to establish regional differences in intensity. Adapted from Golman et al.8

23Na-MRS can be used to produce images based on the sodium content of the myocardium. The regional spectra can be converted into signals producing an MR image of the myocardium, based on sodium content. As the sodium content alters in ischaemic and infarcted tissue, 23Na MRI can be used to detect myocardial infarction and viability.w9 w10

Ischaemic heart disease

Ischaemic heart disease develops when oxygen demands outweigh the supply. In daily practice, coronary angiography (CAG) or coronary artery CT are the standard procedures to rule out significant obstructive coronary artery disease (CAD). However, ischaemia can also occur without significant stenosis. The cause of ischaemia in the absence of significant CAD may be microvascular disease and is associated with a worse prognosis in terms of survival and reinfarction.w11 Furthermore, the discrimination between non-viable (scarred) and viable (hibernating or stunned) myocardium is of clinical importance to determine the potential benefit of revascularisation.

Since oxygen is crucial for the mitochondrial process of oxidative phosphorylation, ischaemic myocardium is largely dependent on inefficient anaerobic glycolysis for the production of ATP. In ischaemic myocardium ATP production, and thus formation of PCr, decreases. A decrease in the PCr/ATP ratio reflects exhaustion of energy reserves of the myocardium.

31P-MRS has been used in extension to investigate cardiac energy metabolism in ischaemic and infarcted tissue. Weiss et al performed a study in patients with >70% stenosis of the left anterior descending coronary artery (LAD) before and after revascularisation. Myocardial ischaemia was provoked by handgrip exercise. Patients with stenosis of the LAD had a significantly decreased PCr/ATP ratio during exercise while no difference between cases and controls could be established at rest. After revascularisation the ‘between group’ and ‘within group’ difference was recovered.3 An interesting finding was that in women admitted with chest pain with normal CAG, seven out of 35 had an abnormal PCr/ATP ratio during handgrip exercise.4 To evaluate the prognostic finding of an abnormal PCr/ATP ratio, a prospective follow-up study was performed in women with chest pain and normal CAG, with or without an abnormal PCr/ATP ratio, and a reference group of 352 women with known CAD. After adjusting for CAD and cardiac risk factors, a PCr/ATP ratio decrease of 1% increased the risk for cardiovascular events by 4%.4 ,5 The most likely explanation for these findings is that patients with normal CAG and abnormal 31P-MRS suffer from microvascular pathology.

As opposed to viable myocardial tissue, the PCr/ATP ratio is not useful in infarcted tissue since both PCr and ATP concentrations are reduced.w12 However, absolute concentrations of PCr and ATP do correlate with infarct size and viability, respectively. In a study in humans with either a fixed or reversible defect on 201Tl scan, myocardial PCr content decreased significantly in patients with reversible and fixed defects when compared to healthy controls. ATP content decreased significantly in subjects with fixed defects, but not in those with reversible defects.6 These data suggest that 31P-MRS could be used to determine myocardial viability. Moreover, in a rat model of myocardial infarction, whole heart PCr content was inversely correlated with infarct size, whereas ATP distribution provided a profile of viable myocardium around the infarction reflecting remodelling of the heart.w13

In rats that underwent ligation of the LAD, 1H-MRS proved lower creatine content as compared to controls.w14 In line with this, in humans total creatine content was significantly lower in regions of infarction.w6

In animal studies it has been shown that viable and non-viable myocardium have a different sodium content as detected by 23Na-MRI. 23Na signal intensity (SI) is elevated in acute ischaemia and remains increased afterwards in humans.w15 To investigate further the role of 23Na-MR imaging in ischaemic heart disease, 23Na-MRI was compared with cine MR imaging and DE imaging in 15 patients with subacute infarction and 15 with chronic infarction. In subacute infarction all patients showed regional elevated 23Na SI that correlated well with wall motion abnormality (WMA) and with DE (r=0.68). In chronic infarction, the patients who had presented with WMA showed increased 23Na SI even in cases without DE, indicating that 23Na MRI could detect hibernating myocardial tissue.7

As shown in figure 4, 13C-MRS has the potential to distinguish between viable and non-viable myocardial tissue in pigs.8 However, this technique has not yet been performed in cardiovascular research in humans.

A major drawback of the application of MRS in general—although significant in ischaemic heart disease—is the low spatial resolution. Since ischaemic heart disease does not involve the whole myocardium, information on localisation of the ischaemic or infarcted area is hard to establish with the current techniques. Overcoming these issues with, for example, higher field imaging or hyperpolarisation could pave the way for MRS to be of diagnostic relevance, most notably in patients with suspected ischaemic heart disease without coronary artery stenosis on CAG. Moreover, MRS could be of use in patients to discriminate between viable and non-viable myocardium.

Heart failure

Heart failure is the final common pathway of a diverse set of specific diseases which will be described in this section. Heart failure in general, its prognosis, and the effect of (non-) pharmacologic interventions have been studied with MRS. The most commonly studied causes of heart failure are dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), and cardiomyopathy related to the metabolic syndrome and T2DM. Metabolic imaging in heart failure has generally focused on two aspects: (1) high energy phosphate metabolism (31P) and (2) substrate utilisation and myocardial TG content (1H).

Decreased ATP metabolism is the final common pathway in heart failure and can be measured by 31P-MRS. Both the absolute level of ATP and ATP flux are diminished.9w16 The PCr/ATP ratio may underestimate the metabolic derangement because both PCr and ATP concentrations are decreased.w17

The causality between altered substrate preference (FFA vs glucose) and heart failure is controversial and beyond the scope of this review. To summarise clinically relevant findings:

  • High concentrations of circulating FFAs are related to cardiac lipotoxicity and reduce cardiac glucose utilisation.10

  • In advanced heart failure systemic and cardiac insulin resistance are related to myocardial lipotoxicity irrespective of the presence of obesity.w18

  • Therapeutic interventions in substrate utilisation can improve heart function. For example, inhibition of fatty acid oxidation and stimulation of glucose use with trimetazidine improved the PCr/ATP ratio and cardiac function in heart failure patients.11

Dilated cardiomyopathy

DCM is characterised by left ventricular dilatation, myocardial wall thinning, and decreased systolic function. Underlying mechanisms are cardiomyocyte death and myocardial fibrosis caused by a heterogeneous group of mutations in different pathways.w19 These mechanisms reflect global cardiac metabolic changes, allowing regional spectroscopy techniques to be representative of energy metabolism in the whole heart. MRS has provided an insight into the metabolism of cardiomyocytes in DCM. The severity of DCM in terms of left ventricular end-diastolic wall thickness and ejection fraction showed a linear regression with the PCr/ATP ratio.w20 Whether metabolic derangements precede heart failure and play a role in the aetiology, or it is merely an innocent bystander, could not be extracted from this observational study. There are, however, indications that alterations in high energy phosphate metabolism play an important role in the pathogenesis of DCM. First, in patients with DCM a decreased PCr/ATP ratio has been shown to be an independent predictor of both total and cardiovascular mortality and offered significant independent prognostic information when compared to New York Heart Association (NYHA) functional class.12 Second, in patients and carriers with Duchenne or Becker muscular dystrophy (the Xp21 dystrophies) the reduced PCr/ATP ratios were not associated with left ventricular mass or ejection fraction. As patients and carriers of the Xp21 dystrophies frequently develop cardiac hypertrophy and DCM, this study suggests that the metabolic derangement precedes cardiac structural changes and failure.w21 Moreover, diuretics and ACE inhibitors induce an improvement in the PCr/ATP ratio parallel to clinical improvement, suggesting that MRS could be used as a quantitative marker of success of therapy.13 In the future 13C-MRS might be used to unravel the causality between substrate use and heart failure as shown in an experimental model of pigs with DCM.w22

Hypertrophic cardiomyopathy

Cardiac hypertrophy has different aetiologies. It can be caused by genetic factors (HCM), it may be the result of chronic hypertension, and it is also seen physiologically in athletes. Metabolism in these different hypertrophic hearts has been studied predominantly with 31P-MRS. In HCM a reduced PCr/ATP ratio has been a consistent finding, in the context of hypertension,14 as well as in young asymptomatic HCM patients.15 In contrast, there is a normal PCr/ATP ratio in physiologic hypertrophy.16 The fact that a reduced PCr/ATP ratio precedes the symptomatology of HCM patients suggests a causative role for metabolic derangement in HCM. Further evidence of this is that in patients with a genetic HCM due to sarcomeric gene mutations, the metabolic derangement was not correlated with the degree of hypertrophy.w23 This study therefore suggests that a decrease in the PCr/ATP ratio is not a consequence of hypertrophy/heart failure/strain but may represent the beginning of the pathophysiological process. The fact that energy deficiency is at least partly responsible for heart failure was shown by a study by Abozguia et al.17 They proved in a randomised, placebo controlled trial that the PCr/ATP ratio increased with perhexiline, an agent thought to improve cardiac energetics by providing a shift from fatty acid oxidation to glucose utilisation. The increased PCr/ATP ratio was associated with an increase in exercise capacity and improvement in diastolic function.17 w23 31P-MRS has the potential to play a role in determining the prognosis of patients with HCM and support intervention trials focused on the potential of improving energy metabolism to ameliorate cardiac function.

Metabolic syndrome

Apart from an increased risk of coronary events, patients with the metabolic syndrome (including T2DM) have an increased risk of developing heart failure without evidence of ischaemia/infarction—the so called diabetic cardiomyopathy. The pathophysiology of cardiac dysfunction in the metabolic syndrome is rather complex and has been reviewed by van der Meer et al.18 It appears that cardiac energy metabolism plays a crucial role in diastolic and systolic dysfunction in these patients. Cardiac energy metabolism in the metabolic syndrome is characterised by an increase of fatty acid uptake, and oxidation becomes less efficient.10 Consistent with this hypothesis, the PCr/ATP ratio was inversely correlated with fasting FFA concentrations in T2DM patients while the PCr/ATP ratio was lower compared to healthy controls, despite the fact that cardiac mass and function appeared to be normal.w24 Accumulation of TGs (as detected by 1H-MRS) in the heart could play a role in the pathogenesis of heart failure in patients with obesity and/or diabetes. There is profound evidence that myocardial TGs reflect cardiomyocyte function by a process called lipotoxicity: the accumulation of toxic lipids such as diacylglycerol and ceramide.w18 w25 Indeed myocardial TG content has been shown to be an independent predictor of diastolic dysfunction.2 Interestingly diastolic dysfunction can be reversed, in parallel with decreased myocardial steatosis, by prolonged caloric restriction in T2DM patients.19 Another study with T2DM patients has proven that a short term increment of FFA flux induced by fasting is associated with myocardial steatosis and diastolic dysfunction.w26 However, in morbidly obese patients undergoing bariatric surgery, the improvement of diastolic function after 6 months was not associated with a change in myocardial TG content.w27 1H-MRS and 31P-MRS were also used to investigate the mechanism of improvement of cardiac function by pioglitazone. Here, the functional improvement was not associated with metabolic changes.w28 In the future MRS could be helpful to unravel the pathophysiology of cardiovascular disease in the metabolic syndrome, and aid in the evaluation of the effect of lifestyle and/or pharmacologic interventions, especially when combined with cardiovascular MRI. 13C-MRS could help to study the complex interplay between substrate use and heart failure in patients with the metabolic syndrome/diabetic cardiomyopathy.20

Valvular heart disease

In valvular heart disease the timing of valve replacement is an important issue: a balanced decision must be taken on the basis of expected benefits versus the risk related to the replacement procedure. Studies involving 31P-MRS of the heart have been performed in patients with aortic stenosis (AoS), aortic incompetence (AoI), and mitral regurgitation. In line with observations in patients with heart failure, the PCr/ATP ratio was reduced in AoS and AoI patients with symptoms of heart failure when compared to asymptomatic patients.w29 A similar study performed by Neubauer et al concordantly showed a decreased ratio only in patients with NYHA functional class III or IV. The ratio was more disturbed in AoS patients with the highest wall stress.w30 In 22 patients with mitral regurgitation the PCr/ATP ratio was higher in those with symptoms and more severe disease. Again, there was a correlation between the PCr/ATP ratio and various parameters of ventricular dysfunction. In this study there was no correlation with wall stress, however.w31 Breyerbacht et al performed 31P-MRS before and 40 weeks after aortic valve replacement in nine AoS patients. PCr/ATP ratios improved significantly from 1.28 to 1.47.w32 The fact that the PCr/ATP ratio is a marker of disease severity in valvular heart disease and is an independent predictor of mortality in DCM suggests that serial 31P-MRS assessments in patients with valvular heart disease could play a role in the timing of a surgical procedure. However, further studies are needed in this regard. In the research setting, MRS could help to unravel the causal relationship between impaired cardiac energy metabolism and mechanical changes: how do they interact to ultimately result in a failing heart?

Heart transplantation

Cardiac allograft patients are at risk for rejection. The gold standard to diagnose rejection is based on histology of the myocardium which warrants repeated biopsy procedures in the post-transplant period.w33 Therefore, a non-invasive tool for this diagnosis is desirable. To investigate the sensitivity of MRS in predicting rejection, 31P-MRS at rest was performed in 14 patients and 17 healthy controls late after transplantation (months to years). The PCr/ATP ratio was significantly lower in transplant recipients than healthy controls. However 31P-MRS could not reliably predict in which patients augmented immunosuppressive therapy was warranted.w34 Van Dobbenburgh et alw35 performed a similar study but assessed 31P-MRS repeatedly in the early post-transplant period to investigate whether 31P-MRS could detect early acute rejection. Again PCr/ATP ratios were lower in heart transplant recipients than controls, with the ratio recovering in time. However, no correlation was found between the PCr/ATP ratios and biopsy scores of rejection: 31P-MRS was unable to predict early acute rejection.w35 Evanochko et alw36 performed a 31P-MRS stress test in heart transplant recipients with normal CAG. Interestingly, 10 out of 25 had a positive stress test with a mean decrease of the PCr/ATP ratio of 25.6±3.6%. Further research is needed to determine the cause of this finding. A possible explanation could be that the reduced ratio is an early sign of chronic allograft vasculopathy.w36 These findings offer hope in regard to implementing MRS techniques in the follow-up of heart transplant patients, although the exact role of MRS remains to be established.

Future direction

The fact that function of the heart depends on myocardial energy metabolism makes MRS an excellent technique to evaluate diseased myocardium. However, the low spectral and temporal resolution combined with the lack of availability of MRS to study the heart in most clinics hampers widespread use in clinical cardiology. Moreover, the currently used techniques remain time consuming.

Whether cardiac spectroscopy will aid in diagnostic and therapeutic decision making depends on the achievement of better spatial and temporal resolution as well as reproducibility. There are various technical opportunities to achieve this goal. First of all, by using higher field MR systems (>3 T), the advantage of increased SNR could result in better resolution and reproducibility.w37 Furthermore, development of improved shimming algorithms, pulse sequences, and receiver coils could aid in achieving higher spatial resolution MRS—that is, to be able to perform an MRS based analysis in line with the 17-segment model for detection of regional metabolic alterations specific for (non) viable myocardium.

One of the most promising new techniques is hyperpolarisation. Hyperpolarised 13C with various tracers can assess substrate flux and thereby gain a position in clinical practice.20 w38


MRS is a unique, non-invasive, non-irradiating tool to assess cardiac metabolism in humans. In the research setting, MRS has provided insight into the pathophysiology of multiple cardiac diseases. The tight relation between energy metabolism and cardiac function potentiates spectroscopy to be of use in diagnostics, prognosis, and treatment guidance in cardiology practice. Continuous improvements regarding technical optimisation, and exciting new techniques such as hyperpolarisation, may help to implement cardiac MRS in selected patients in daily practice.

Cardiac magnetic resonance spectroscopy: key points

  • Cardiac magnetic resonance spectroscopy (MRS) is a unique non-invasive, non-irradiating tool to assess cardiac energy metabolism in vivo in health and disease.

  • Different nuclei can be used for the evaluation of metabolism with MRS: 1H, 31P, 13C, and 23Na.

  • In the cardiomyocyte free fatty acids and glucose are the main suppliers of chemical energy necessary for the generation of ATP in mitochondria. ATP is necessary for both systolic and diastolic function.

  • Chemical energy is transferred from the mitochondrion to the myofibrils by the creatine kinase shuttle.

  • A decreased PCr/ATP ratio as determined by 31P-MRS predicts an increased risk for cardiovascular events in women with chest pain without significant stenosis on coronary angiography.

  • The failing heart has a decreased PCr/ATP ratio which in dilated cardiomyopathy is an independent predictor of mortality.

  • Myocardial triglyceride content as determined by 1H-MRS has been shown to be an independent predictor of diastolic dysfunction in the ageing and diabetic heart.

  • The future application of cardiac MRS in clinical practice is reliant on crucial technical improvements that generate higher spatial resolution data.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

  • RCP credits: Log your activity in your CPD diary online (—pass mark is 80%.

  • EBAC credits: Print out and retain the BMJ Learning certificate once you have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits can now be converted to AMA PRA Category 1 CME Credits and are recognised by all National Accreditation Authorities in Europe (

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.


We thank G Kracht for support on figures and table.


  1. This paper reviews the interrelationship between disruptions in cardiac energetics and heart failure.
  2. The first study in humans assessing high energy phosphate metabolism in myocardial ischaemia in vivo.
  3. A detailed review on the pathophysiology of cardiac dysfunction in obesity.
  4. This study signifies the role of MRS to monitor the strong association between cardiac energetics and function.
  5. The power of MRS in determining the prognosis of DCM patients indicates the potential benefit of MRS in clinical cardiology.
  6. The first report on the effect of treatment on cardiac high energy phosphate metabolism.
  7. This study shows that MRS is a useful tool for research on the pathophysiology of cardiomyopathy.
  8. The clinical relevance of cardiac energetics is reflected by the potential of metabolic therapy to improve heart function.
  9. A detailed description of the role of MRS and MRI in the metabolic syndrome.
  10. An exciting review on the clinical potential of hyperpolarised 13C-MRS.
View Abstract


  • Contributors All authors contributed to the search of the literature, text editing and composition of figures and tables.

  • Competing interests In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The authors have no competing interests.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles