- Email: [email protected]
MRI
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS European Journal of Radiology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Imaging of the myocardium using 18 F-FDG-PET/MRI Jiˇrí Ferda (M.D. Ph.D Prof.) a,∗ , Milan Hromádka (M.D. Ph.D) b , Jan Baxa (M.D. Ph.D) a a b
Clinic of the Imaging Methods, University Hospital Plzen, Alej Svobody 80, 304 60 Plzen, ˇ Czech Republic Department of Cardiology, University Hospital Plzen, Alej Svobody 80, 304 60 Plzen, ˇ Czech Republic
a r t i c l e
i n f o
Article history: Received 23 March 2016 Received in revised form 17 July 2016 Accepted 19 July 2016 Keywords: Myocardium Positron emission tomography PET/MRI Fluorodeoxyglucose
a b s t r a c t The introduction of the integrated hybrid PET/MRI equipment creates the possibility to perform PET and MRI simultaneously. Depending on the clinical question, the metabolic conversion to glycolytic activity or beta-oxidation is performed before the application of FDG. Since FDG aids to evaluate the energetic metabolism of the myocytes and myocardial MRI reaches the imaging capabilities of perfusion and tissue characterization in the daily routine, FDG-PET/MRI looks to be a promising method of PET/MRI exploitation in cardiac imaging. When myocardial FDG uptake should be evaluated in association with the perfusion distribution, the cross-evaluation of FDG accumulation distribution and perfusion distribution pattern is necessary. The different scenarios may be used in the assessment of myocardium, the conversion to glycolytic activity is used in the imaging of the viable myocardium, but the glycolytic activity suppression might be used in the indications of the identification of injured myocardium by ischemia or inflammation. FDG-PET/MRI might aid to answer the clinical tasks according to the structure, current function and possibilities to improve the function in ischemic heart disease or to display the extent or activity of myocardial inflammation in sarcoidosis. The tight coupling between metabolism, perfusion and contractile function offers an opportunity for the simultaneous assessment of cardiac performance using one imaging modality. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hybrid imaging has brought a powerful tool to clinical practice that combines several advantages of both morphological and metabolic imaging. During the last two decades, 18 F-fluorodeoxyglucose (FDG) has changed the clinical spread of positron emission tomography (PET) imaging like no other molecule. The universal and achievable behaviour of FDG predicts it to be widely available every day and everywhere. Since its first years, PET/CT using FDG has played a crucial role in oncological imaging. Due to the technical difficulties in the co-existence of the PET acquisition system with the high magnetic field, the clinical use of PET/MRI was delayed for about fifteen years. Oncological imaging using FDG also still looks dominant in the current indications of PET/MRI. The introduction of the integrated hybrid PET/MRI equipment creates the possibility to perform PET and MRI simultaneously. The advent of this new technique has introduced the idea of new concepts in myocardial imaging [1,2], respecting all contraindications of the MRI in patients with incompatible implants
∗ Corresponding author. E-mail addresses: [email protected] (J. Ferda), [email protected] (M. Hromádka), [email protected] (J. Baxa).
or claustrophobia. As magnetic resonance is the gold standard in vivo technique in the assessment of the myocardial structure, positron emission tomography has become the gold standard of the in vivo investigation of its metabolism. The synergy in myocardial imaging should be the driving force of myocardial PET/MRI concept. Due to the advantage of the choice, the selection of the most effective way to assess myocardial perfusion, metabolism and function has to be evaluated and the possible role of myocardial PET/MRI has to be claimed. Since FDG aids to evaluate the energetic metabolism of the myocytes [3] and myocardial MRI reaches the imaging capabilities of perfusion and tissue characterization in the daily routine, FDG-PET/MR looks to be a promising method of PET/MRI exploitation in cardiac imaging, even if other tracers like 13 N-ammonium, 15 O-water, 82 Rb-rubidium-chloride, 18 F-flurpiridaz or 11 C-palmitate may be used in cardiac PET.
2. Myocardial energy metabolism and FDG uptake A major role of FDG use in PET myocardial imaging is played by the relationship between -oxidation or oxidative glycolysis based energy metabolism [4]. Myocardial energy metabolism depends on the oxidation of various substrates. Under fasting conditions, the myocardium uses almost all energy for the production of adenosine
http://dx.doi.org/10.1016/j.ejrad.2016.07.014 0720-048X/© 2016 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
2
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
triphosphate (ATP) during the oxidation of free fatty acids [5]. Fasting decreases plasma glucose levels rapidly, and the free fatty acids levels become high. The extraction of fatty acids by the myocardium is enhanced, and long-chain acyl-coenzym A (CoA) is increasingly formed. Even the activated fatty acids could be used in triacylglycerol or phospholipids building, as fasting facilitates their role in energy metabolism. Fatty acid energy-based metabolism takes place inside the mitochondrion. Acyl-CoA is useful as a source of substrate of -oxidation and the majority is used in ATP synthesis. However, the molecule of acyl-CoA is too large to be transported into the mitochondria, the carnitine shuttle aids to pass the longchain fatty acids with more than 15 acetate parts. The product of the -oxidation acetyl-CoA then enters the cycle of Krebs (tricarboxylic acid cycle), the final pathway of the oxidative metabolism of all substrates. When the insulin plasma level is low and free fatty acids levels increase, only a small amount of the glucose is extracted by the myocardium [6,7]. However, when the glucose level increases, the energetic role of the oxidative glycolysis becomes more important. In a postprandial state, one-third to one-half of energy used by the myocardium depends on oxidative glycolysis [8]. Beyond the physiological conditions, the state after myocardial ischemia is the situation in which the up-regulation of the oxidative glycolysis becomes the one-way energetic pathway [9–11]. Nonetheless, not only the acute situation of the ischemia enhances the glycolysis; the glucose metabolism also remains for several weeks after the ischemic episode, partially due to inflammatory cell immigration into the insulted myocardium, partially when the myocardium is hibernating, even if the myocardium is recovering in normal perfusion conditions and functional ability (Fig. 1). The FDG molecule looks very similar to that of glucose, only the OH group is replaced by the fluorine. A very similar molecular appearance is responsible for FDG behaviour [3]. FDG accumulates in the myocardial tissue proportionally according to the glucose transport and phosphorylation, and even FDG exhibits a slightly different affinity to transporters and phosphatase compared to glucose. The utilization of the glucose does not have a uniform pattern in the entire heart, and, consequently, the inhomogeneous FDG uptake distribution is obvious [8]. Copying the glucose metabolism, the lateral wall of the left ventricle exhibits a more homogenous and more intense distribution of FDG [9,12]. Because of the outlined variability of the energetic sources, the major role in myocardial metabolic conversion belongs to the patient’s dietary preparation and pharmacological intervention. As during fasting conditions, the glucose levels are decreasing, therefore the metabolism of the myocardium converts towards fatty acid -oxidation. This fact makes the PET images after intravenous administration of FDG poor and of unstable quality [13]. However, the fasting state before the imaging has been applied to detect the ischemic myocardium after physical exercise. Fasting helps to identify myocardial infarction, infarcted tissue exhibits an increased level of glucose metabolism not only due to the myocardial glucose conversion, but also due to the migration of inflammatory vectors such as macrophages and/or activated fibroblasts. To be able to delineate the normal myocardium, it is necessary to make its glucose metabolism as low as possible. An effective way to suppress glycolytic activity in normal myocytes is a low carbohydrate and high fat diet (the Atkins diet) [14]. There is also an alternative way to myocardial conversion to fatty acid exploitation. The intravenous administration of un-fractioned heparin leads to rapid conversion to -oxidation in the myocardium. Having a direct impact on lipoprotein lipase activation, the heparin causes the elevation of the plasma free fatty acids levels and switches the normal myocytes almost exclusively to -oxidation [15]. The metabolic switch almost exclusively to glucose is the important challenge in the assessment of myocardial viability. The main strategy is to offer to the myocardium such a glucose overload
that myocytes are able to diminish the -oxidation and the forced glucose extraction results in high FDG uptake. There is a traditional approach and an alternative one. Some centres advocate the hyperinsulinemic-euglycemic clamp – to intravenously administer insulin and glucose simultaneously under control of normal glucose plasma levels. The main intent is to standardize the glucose metabolism within the whole myocardium, but the most important disadvantage is the complicated procedure [16]. An easier way to convert the myocardium to glycolysis is the oral glucose load. Oral glucose administration increases plasma insulin level and accelerates the glycolytic activity and decreases the plasma fatty acid level. Oral glucose load very effectively forces the conversion in non-diabetic patients. In diabetics, there are two main problems in myocardium glycolytic conversion. High glucose plasma levels lead to the limited ability of the myocardium to extract FDG. The second very important problem could be the insensitivity of some diabetics to insulin, which has a key role in metabolic switching [6,7]. The classic approach of the FDG uptake evaluation is based on proportional assessments according to myocardial perfusion [16,17]. Proven concepts include the perfusion evaluation using PET tracers such as 15 O-H2 O, 13 N-NH3 or 82 Ru. PET-based perfusion imaging could be successfully replaced by SPECT imaging using 99m Tc labelled methoxy-lisobuthyl-isonitril or 201 Tl. However, cardiac magnetic resonance offers perfusion imaging based on the first pass of the gadolinium contrast agent (Fig. 4).
3. Myocardial gadolinium imaging during PET/MRI Gadolinium-based contrast agents have been used for thirty years to involve the tissue behaviour in T1 weighted images (with a known influence also in T2 weighted images). Gadolinium based contrast agents (GBCA) have also been used in myocardial imaging for the past two decades, and the possibilities of myocardial imaging using magnetic resonance and the GBCA application have been studied intensively. Concerning the myocardium, multiparametric ECG synchronized cardiac magnetic resonance imaging (CMRI) included kinetic studies, the first past perfusion imaging and the evaluation of the myocardial structure, including the late gadolinium enhancement evaluation. Using the basic sequences, the associate findings such as pericardial fluid or valvular disease could be found at the same time. With respect to specific questions like flow quantification, valve regurgitation or contractile myocardial function, the sequences targeting blood flow (phase contrast imaging) or myocardial contractility (myocardial tagging) may be used. Multiple studies assessed the clinical relevance of flow-limiting coronary artery disease (CAD). CMRI detected more cases of obstructive coronary artery disease without increasing the falsepositive rate; however, the sensitivity of nuclear stress testing is significantly lower than had previously been published. Very few centres now use CMRI for the diagnosis of CAD given the lack of experienced clinicians able to perform the study, the inability to perform exercise stress CMRI, and patient issues including claustrophobia and relatively long imaging times. Nevertheless, studies were able to show that CMRI can serve as an alternative to stress echocardiography and nuclear stress testing when evaluating patients for obstructive CAD (Fig. 7). When myocardial FDG uptake should be evaluated in association with the perfusion distribution, the cross-evaluation of FDG accumulation distribution and perfusion distribution pattern is necessary. Modelling of the myocardial flow using GBCA is completely different than that used in PET. PET flow modelling is based on indicator extraction by cardiomyocytes, so it uses intracellular uptake of the tracer. Because neither the time resolution nor
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
3
Fig. 1. A male patient after acute myocardial infarction and consequent angioplasty of the left anterior descending artery, images in short axis view. Multiparametric assessment of the myocardium: (a) – early phase of the first pass perfusion using turboFLASH T1 sequence, (b) – late phase of the first pass perfusion using turboFLASH T1 sequence, (c) – late gadolinium enhancement image using phase sensitive inversion recovery sequence, (d) – perfusion analysis – maximum slope map representing the perfusion flow function, (e) – perfusion analysis – initial area under the curve map representing the flow volume, (f) – FDG distribution on PET image representing the level of glycolytic activity. Within the lateral wall of the left ventricle, there is the hypo-perfusion zone with sub-endocardial infarction exhibiting the sub-endocardial late enhancement, the metabolism is diffusely decreased in the same region.
Fig. 2. A male patient one year after large myocardial infarction in the interventricular septum suffering from low left ventricle ejection fraction of 20%, images in short axis view, patient prepared with oral glucose load, PET/MR fusion: (a) – pure FDG-PET image, (b) fused FDG-PET/MR image, (c) – late gadolinium enhancement MR image. The infarction of the septum contains the islets of the tissue without LGE, but accumulating FDG. The tissue exhibits the viable population of the myocytes.
the spatial resolution is sufficient, no first pass imaging is possible. Magnetic resonance imaging is traditionally based on first pass perfusion; it exploits the T1 weighted steady state sequences such as turboFLASH. Other MRI related perfusion techniques such as gadolinium-based T2* perfusion, arterial spin labelling (ASL) and BOLD (blood oxygen level dependent) imaging have not been widely distributed. When three Tesla systems enable high resolution, the steady state T1 weighted images might be used by hybrid integrated PET/MRI equipment [18,19]. Depending on the heart rate of an individual, three to six planes could cover the myocardium. The first pass of the contrast agent through the coronary riverbed takes about one breath hold, thus the perfusion data set could be acquired within one sustained breath hold (Fig. 10).
GBCA perfusion imaging tracks the advent of the indicator into the micro-vascular extracellular space of the myocardium and then the evaluation of the saturation of the myocardium during one perfusion cycle [20,21]. When the signal of the myocardium depends on the own properties of the used acquisition sequence, the longitudinal relaxation of the myocardium (related to the water and protein content) and the behavioural properties of the contrast agents (their concentration and/or relaxivity), the quantitative computation of the perfusion within the myocardium might be difficult, even without respect to the movements of the myocardium. Nevertheless, the FDG uptake quantification and standardization within the myocardium is disputable, too. The quantification of the perfusion depends on several factors: the fashion of the bolus appli-
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
4
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
Fig. 3. The same patient like in Fig. 2, MR first pass perfusion: (a) – early phase, (b) mid-phase, (c) – later phase. The septum contains the tissue with preserved perfusion, corresponding to that showed by FDG uptake.
Fig. 4. The same patient like in Figs. 2 and 3, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – anterior wall shows preserved normal perfusion (red), lateral wall with moderate decrease of flow (yellow), inferior wall (green) and septum (blue) suffer from decreased perfusion. Comparison of the perfusion and metabolic patterns enables to distinguish between hibernating myocardium (inferior wall) and scar (septum) with the additional information about islets of viable tissue within the septum. The coronary bypass graft surgery improved the perfusion conditions so, that the left ventricle ejection fraction increased to 35%.
Fig. 5. A male patient five days after large myocardial infarction after occlusion of the left anterior descending artery with akinetic apical third of the left ventricle, images in short axis view, patient prepared with intravenous administration of heparin, PET/MR fusion: (a) – pure FDG-PET image, (b) fused FDG-PET/MR image, (c) – late gadolinium enhancement MR image. The infarction exhibits subendocardial localization with late gadolinium enhancement, present also intraventricular thrombus, the tissue which has injured by the ischemia exhibits extreme FDG uptake, it represents “ischemic memory” of the myocardium.
cation, the concentration and the relaxivity of the used agent, and last but not least the computation model [22]. 15 O-water has to be established as the gold standard perfusion indicator with the linear relation of the extraction independent of the blood flow. All other radiotracers suffer from decreasing extraction in higher flow rates in the progressive row 18 F-flurpiridaz, 13 N-ammonium, 82 Rb-rubidium-chloride [22]. Some opinions anticipate the incomparable perfusion assessment between gadolinium first pass and nuclear medicine methods, but the results of the larger studies and especially results of meta-analyses have confirmed the
clinical relevant results of CMRI perfusion quantification. The perfusion quantification computation model presents equal results independently to the algorithm, even if the model is linear (Fermiconstrained deconvolution), model-independent deconvolution, or one-compartment model or an uptake model (similar to the models of radiopharmaceutical extraction) [22]. The most important advantage of the GBCA is that fact that it acts twice. First pass perfusion is followed by the period of late gadolinium enhancement (LGE) after about ten minutes. The LGE phenomenon is based on two different mechanisms. First, the GBCA
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
5
Fig. 6. The same patient like in Fig. 5, MR first pass perfusion: (a) – early phase, (b) mid-phase, (c) – later phase. The wall contains the tissue with preserved perfusion, except the thin inner rim of subendocardial infarction.
Fig. 7. The same patient like in Figs. 5 and 6, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – non infarcted wall shows preserved normal perfusion (red), zone of subendocardial infarction shows decreased flow (green), the tissue with FDG uptake represents stunned myocardium, the contractility recovered after revascularization.
Fig. 8. A female patient with multiorgan involvement by the sarcoidosis, patient prepared with low carbohydrate high fat diet, PET/MR fusion in horizontal axis: (a) – late gadolinium enhancement MR image, (b) fused FDG-PET/MR image, the FDG uptake is copying the presence of the late gadolinium enhancement and showing the sites of the active inflammation, the maximum activity showed the lateral wall.
kinetics are much slower, and second, the volume of distribution is much higher in the infarcted myocardium. LGE is visible due to the different distribution and release of the gadolinium from the interstitial tissue in the conditions of normal myocardium, myocardium affected by the ischemic necrosis or scar tissue [23]. Nevertheless, the indicator delayed kinetics is the basic principle aids to distinguish between myocardium having its integrity from that which is deteriorated by the inflammation and/or fibrosis dur-
ing acute myocarditis or granulomatous inflammation [23,24]. LGE imaging is traditionally based on approach to use the inversion recovery technique to nullify the signal of the normal myocardium and to make visible gadolinium enhancement, typically using 3D-IR-turboFLASH sequences, or to use the phase sensitive inversion recovery (PSIR). The reconstruction of the phase sensitive images improves splitting the signal of the normal myocardium and gadolinium to the opposite points [24]. On three Tesla systems,
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
6
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
Fig. 9. The same patient like in Fig. 8, PET/MR fusion in horizontal axis: (a) – late gadolinium enhancement MR image, (b) fused FDG-PET/MR image, the FDG uptake is copying the presence of the late gadolinium enhancement and showing the sites of the active inflammation, note the subendocardial involvement of the septum, but also the extreme finding of the interlobular fibrosis within the liver affected by the sarcoidosis too.
Fig. 10. The same patient like in Figs. 8 and 9, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – the perfusion within the FDG accumulating tissue affected by the active sarcoidosis is decreased similarly like in those conditions seen in myocardial ischemia cases.
the contrast-to-noise ratio (CNR) could be improved without the loss of diagnostic accuracy when PSIR is used [25]. The PSIR singleshot technique allows for imaging a set of slices covering the entire myocardium during a single breath hold without adaptation of the inversion time. 4. Myocardial FDG-PET/MRI acquisition protocol and data evaluation Depending on the clinical question, the metabolic conversion to glycolytic activity or beta-oxidation is performed before the application of FDG. Due to long-lasting data acquisition enabling the use of a sufficient number of coincident events, the FDG dosage could drop to 2.5 MBq/kg of body weight. The optimal accumulation time ranges between 45 and 90 min after the actual beginning of FDG intravenous administration [16,25–27]. Since the heart is a continually moving structure, proper data acquisition is crucial for obtaining valuable results. Both PET and MRI acquisition is driven by the electrocardiogram, the prospective triggering is used in PET data acquisition, and alternating prospective acquisition triggering and retrospective data segmentation are used in the MRI part. Due to breathing, breath-navigated acquisition is combined with ECG in the dual-triggering method. Because of the fundamental importance of the attenuation correction (AC), the two point Dixon T1 spoiled gradient echo sequences (VIBE – volume interpolated breath hold examination) followed immediately after the planning sequences are completed [27–29]. With the start of the AC sequence, the PET acquisition is automatically initiated, and the time of the PET data acquisition
might be selected according to the scanning time of the whole CMRI protocol. In between PET data acquisition of kinetic sequences occur in four chamber, longitudinal two chamber and left outflow two chamber views using steady state free precession sequences (trueFISP) or ultrafast gradient echo sequences (turboFLASH). As an option, myocardial contractility may be assessed in which the tagged gradient echo images are performed before the application of the dynamic post-contrast T1 imaging. Myocardial first pass perfusion sequences are based on T1 weighted turboFLASH sequences repeated through the cardiac cycle during the application of the contrast material bolus. Dosed by 0.5 mmol/kg, preferable are highconcentration (gadobutrol) or high-relaxivity agents (gadobenate dimeglumine) providing the rapid slope of the blood signal. Waiting for the late gadolinium enhancement for ten minutes, the short axis sequences covering the whole heart are scanned. LGE images in diastolic phase complete the protocol, and IR-turboFLASH or PSIR sequences are useful as outlined above. The whole protocol lasted 20 min. The sequences with parameters are included in the table. When data acquisition is completed, the PET reconstruction is started. The tissue model (-maps) based on fat and water image derived from opposed-phase and in-phase is created for use in attenuation corrected image reconstruction [27–29]. Possible problems could occur during the presence of the metal implants resulting in correction failure thanks to signal voids on -maps. Nevertheless, these problems don’t often occur even if cerclage of the sternum is present. The algorithm of the ordered subset with maximization expectation (OSEM) in zoomed field of view might be recommended. If the point spread function is used, the increased noise could be achieved, but use of the reconstruction algorithm
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
improve the spatial signal resolution. In order to be comparable with LGE MRI images, the AC PET images are reconstructed in the diastolic phase. The data evaluation is based on a multi-parametric view of the dedicated software platform. The cardiac perfusion might be assessed using the computation of the perfusion maps related to the 17-segment model recommended by the American Heart Association (AHA). The slice-to-slice fusion reordering of the AC PET images with the morphological MRI images helps compare the level of FDG uptake with the structural changes of the myocardium on CMRI – thickness, contractility, hypo-perfusion and/or delayed enhancement. The calculation of the cardiac function including left ventricle ejection fraction, stroke volume and cardiac output are needed to complete the entire information about the myocardium. 5. Current reasonable FDG-PET/MRI imaging strategies 5.1. Viable myocardium Being the classical indication of both worlds, imaging of the viable myocardium may become the crucial indication of hybrid PET/MRI with FDG, even if both methods are overlapping in some aspects of the imaging. When the clinical question is to estimate the possible improvement of the ventricular function after planned revascularization, the detection of the hibernating myocardium is the target of the imaging. The hibernating myocardium survives in conditions of saturation by oxygen and energy substrates on the edge between a sustainable level of basal metabolism and ischemia leading to necrosis. Usually the flow limiting central stenosis causes hypoperfusion and it tears the myocytes apart from sufficient supply by the blood stream, and consequently the myocardium is left in chronic hypoxia. As outlined above, the hypoxia is the driving force in myocardial conversion to glycolytic metabolism. In order to prevent the problems in heterogeneity of energy substrates exploitation, the glycolytic conversion of the normal myocardium is also needed. For the purpose of providing the viable myocardium the highest level of FDG uptake, the patient has to be prepared with fasting with subsequent glucose load, either by oral load or using the hyperinsulinemic-euglycemic clamp. By glucose driven myocardial metabolism might warrant that no viable myocytes population remained undetected. The conventional approach in the evaluation of FDG accumulation within the myocardium demands comparison with the perfusion pattern. As for providing the useful information, first pass gadolinium MR perfusion completes the information. When PET imaging suffers from insufficient spatial resolution to distinguish between thinned myocardium and partially infarcted as in the sub-endocardial layer, LGE MRI supplements this disadvantage with its excellent ability to discriminate scar regions from those that are structurally normal. FDG-PET/MRI might help in fulfilling all the clinical tasks according to the structure, current function and possibilities to improve the function. 5.2. Myocardial ischemia and post-infarction imaging Complete deregulation of the glycolytic energy metabolism in hypoxic myocardium is the leading idea in the direct imaging of the hypoxic or ischemic myocardium [30,31]. When occurring, suddenly myocytes are turned to the pure glycolysis, no available oxygen makes the anaerobic glycolysis the leading way of the energy metabolism. The conversion occurs rapidly in several minutes (even when induced by stress testing). When ischemia occurs, glycolysis remains and lasts after the hypoxic/ischemic period for several days to weeks, even if the blood supply is restored. The up-regulation of glycolysis by the myocytes is one mecha-
7
nism of the increased FDG accumulation within the post-ischemic myocardium; on the other hand, the ischemic necrosis stimulates the migration of the cellular vectors into the affected tissue. Like induced inflammation with monocytes, activated lymphocytes, activated fibroblasts and progenitor cells are a major cause of the increased glucose metabolism inside the infarcted zone forming scar tissue. These inflammatory cells exhibit extremely high levels of glucose transporters and hexokinase activity. In such a case, thorough glycolysis suppression has to be performed. Intravenous un-fractioned heparin administration is the best way to complete metabolic conversion. The dose of 50 IU of heparin per kilogram of body weight may be recommended. Having some indication problems in subacute myocardial infarction, the Atkins diet appears more problematic to use. The evaluation relies on a direct comparison with first pass MRI perfusion and LGE. The gadolinium late stain shows zones of inflammatory tissue response and scar forming, the restored perfusion without late enhancement represents those zones remembering the ischemia. Although stress-induced hypoxic myocardium imaging is not yet clinically established, the infarction-related changes and ischemic memory of the myocardium could be routinely used [32]. Owing to the unsuspected problems in preparation (fasting, glucose overload or heparin application), both with limited experience with myocardial behaviour in stress related to glucose extraction in humans, the further testing of myocardium imaging has to be performed and critically compared to other stress testing methods [18,29,33,34]. Despite the first reports in the assessment of hypertrophic cardiomyopathy [35], there are no clear advantages of PET/MRI versus MRI. 5.3. Myocardial inflammation The inflammatory cellular population reaches one of the highest grades of glucose consumption according to the up-regulation described in the previous paragraph in cases of post-infarction inflammatory response. In its own sense, myocardial inflammatory diseases represent the migration of the activated cells – either monocytes, activated lymphocytes or neutrophils in infectious or para-infectious acute myocarditis, or histiocytes-related elements, such as epitheloid or Langerhans polynuclear cells in granulomatous diseases represented by sarcoidosis [36,37]. The acute onset of myocardial infiltration is linked to the elevation of myocardial biomarkers, especially troponin. Myocardial injury not related to ischemia is one of the most frequent indications of CMRI. To achieve the optimum effectiveness of the discrimination between nonaffected myocardium, active inflammation and non-active fibrous scar tissue, the conversion to -oxidation is needed [16,36,37]. Glycolysis suppression in healthy myocardium and late gadolinium images makes the comparison of the metabolism and structure possible. Having late gadolinium enhancement, the FDG avid tissue represents active inflammation. On the other hand, low accumulation confirms non-active fibrosis in the zone of late gadolinium enhancement. The normal myocardium remains non-enhanced and non-accumulating FDG. 6. Future perspectives of myocardial imaging using FDG-PET/MRI The future wider acceptance of the FDG-PET/MRI as an imaging tool for myoacardial assessment is depending on the future distribution of the equipment. For the future development of indications is very important advantage radiation dose burden to the patient during FDG-PET/MRI. The use of 2.5 MBq/kg in 70 kg patient is related to the dose about 3,5 mSV, that dose is similar to the abdominal CT examination.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
8
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
When the myocardial injury related or not related to ischemia is still increasing indication of CMRI, the imaging of the activity of the inflammation could plays a role in the prognosis estimation in patients with wide range of causes of myocardial impairment – acute ischemia, myocarditis or sarcoidosis. More precise detection of the process extent and its activity using a combination of the T1 and T2 mapping, water molecule diffusion and perfusion imaging might with imaging of the metabolically active nonmyocyte-cellular population. In the near future, that combination of multiparametric imaging of the metabolism, microstructure and microvascular perfusion could aid to achieve the optimum effectiveness of the discrimination between non-affected myocardium, active inflammation and non-active fibrous tissue. Simple and quick conversion to -oxidation using non-fractioned heparin makes the direct inflammation imaging easier, so the increased role of the metabolic suppression of the healthy myocardium could be awaited. Even the original indication of FDG-PET was the imaging of the viable myocardium, the future use will be further in the shadow of LGE. And what about the imaging beyond fluorodehydroxyglucose? Besides the FDG imaging, the other radiopharmaceuticals are able to play some role in myocardial imaging. The hypoxia imaging has some potential in the estimation of the hibernating myocyte population, the promising should be fluoromisonidazol of 64 Cu labelled substances. The special requests for amyloid imaging within myocardium maybe fulfil the drugs similar like florbetaben or flutemetamol, whose are used for brain amyloid imaging. The investigation of the neural loss in some patients could be performed using fluorodihydroxyphenylalanine. But the spread of these methods will last more time, because of needed effectivity and safety studies. For a closer future, the myocardial PET/MRI will still remain the FDG based.
7. Conclusion As FDG is a common radiopharmaceutical used in PET imaging and as there is long experience with CMRI, the natural combination of both methods in PET/MRI became the first choice in the introduction into clinical practice. Metabolic imaging combined with excellent structural information, even more supplemented with perfusion imaging using MRI is one of the major strengths. Also there are no problems with short-living isotopes needing on-site cyclotrons. While the sophisticated viability assessment looks like the easiest way to clinical adoption, the hypoxic/ischemic memory or infarction extend assessments need clinical relevant questions [18,19]. The final advantage, the lower radiation dose than is used in one chest CT, makes PET/MRI preferable to PET/CT, when the results of PET remain comparable [38]. However, while myocardial FDG-PET/MRI is promising, the other kinds of radiopharmaceuticals remain in the shadows. Other radiopharmaceuticals are widely useful, some are still not registered (18 F-flurpiridaz in perfusion imaging), registered in other indication (18 F-florbetaben in amyloid imaging, 18 F-fluorodihydroxyphenyalanine in dopaminergic imaging), closely related to cyclotron availability (13 N-ammonium, 15 O-water or 11 C-palmitate), or bearing the high radiation dose burden to the patient (82 Rb-rubidium-chloride). In the coming years, further evolution of the method might be observed.
Conflict of interest The authors have no conflict of interest to disclose.
Authors contribution We wish to confirm there has been no significant financial support for this work that could have influenced its outcome. All authors have read and approved submission of this manuscript. This work has not been published and is not being considered for publication elsewhere. Acknowledgements Supported by project Conceptual development of the research institution of the Czech Ministery of Health No. 00669806 – FN ˇ and by Program of the development of the Charles´ı University Plzen (project P36). References [1] C. Rischpler, S.G. Nekolla, K.P. Kunze, M. Schwaiger, PET/MRI of the heart, Semin. Nucl. Med. 45 (3) (2015) 234–247. [2] C. Rischpler, S.G. Nekolla, I. Dregely, M. Schwaiger, Hybrid PET/MR imaging of the heart: potential, initial experiences, and future prospects, J. Nucl. Med. 54 (3) (2013) 402–415. [3] A.J. Liedtke, B. Renstrom, S.H. Nellis, Correlation between [5-3H]glucose and [U-14C] deoxyglucose as markers of glycolysis in reperfused myocardium, Circ. Res. 71 (3) (1992) 689–700. [4] A.J. Liedtke, Alterations of carbohydrate and lipid metabolism in the acutely ischemic heart, Prog. Cardiovasc. Dis. 23 (5) (1981) 321–336. [5] P.1 Camici, E. Ferrannini, L.H. Opie, Myocardial metabolism in ischemic heart disease: basic principles and application to imaging by positron emission tomography, Prog. Cardiovasc. Dis. 32 (3) (1999) 217–238. [6] D.1 Sun, N. Nguyen, T.R. DeGrado, M. Schwaiger, FC 3rd. Brosius, Ischemia induces translocation of the insulin-responsive glucose transporter GLUT4 to the plasma membrane of cardiac myocytes, Circulation 89 (2) (1994) 793–798. [7] S.1 Egert, N. Nguyen, M. Schwaiger, Myocardial glucose transporter GLUT1: translocation induced by insulin and ischemia, J. Mol. Cell. Cardiol. 31 (7) (1999) 1337–1344. [8] R.C. Marshall, S.C. Huang, W.W. Nash, M.E. Phelps, Assessment of the [18F]fluorodeoxyglucose kinetic model in calculations of myocardial glucose metabolism during ischemia, J. Nucl. Med. 24 (11) (1983) 1060–1064. [9] G. Heusch, The regional myocardial flow-function relationship: a framework for an understanding of acute ischemia, hibernation, stunning and coronary microembolization. 1980, Circ. Res. 112 (12) (2013) 1535–1537. [10] G. Heusch, R. Schulz, S.H. Rahimtoola, Myocardial hibernation: a delicate balance, Am. J. Physiol. Heart Circ. Physiol. 288 (3) (2005) H984–999. [11] L.H. Opie, Effects of regional ischemia on metabolism of glucose and fatty acids Relative rates of aerobic and anaerobic energy production during myocardial infarction and comparison with effects of anoxia, Circ. Res. 38 (5 Suppl. 1) (1976) 152–174. [12] A.J. Liedtke, H.C. Hughes, J.R. Neely, Effects of excess glucose and insulin on glycolytic metabolism during experimental myocardial ischemia, Am. J. Cardiol. 38 (1) (1976) 17–27. [13] R.J. Gropler, B.A. Siegel, K.J. Lee, S.M. Moerlein, D.J. Perry, S.R. Bergmann, E.M. Geltman, Nonuniformity in myocardial accumulation of fluorine-18-fluorodeoxyglucose in normal fasted humans, J. Nucl. Med. 31 (11) (1990) 1749–1756. [14] Y. Kobayashi, S. Kumita, Y. Fukushima, K. Ishihara, M. Suda, M. Sakurai, Significant suppression of myocardial (18)F-fluorodeoxyglucose uptake using 24-h carbohydrate restriction and a low-carbohydrate, high-fat diet, J. Cardiol. 62 (5) (2013) 314–319. [15] A.M. Scholtens, H.J. Verberne, R.P. Budde, M. Lam, Additional heparin pre-administration improves cardiac glucose metabolism suppression over low carbohydrate diet alone in 18F-FDG-PET imaging, J. Nucl. Med. (December) (2015), pii:jnumed.115.166884. [Epub ahead of print] PubMed PMID: 26659348. [16] A.1 Klaipetch, O. Manabe, N. Oyama-Manabe, S. Chiba, M. Naya, S. Yamada, K. Hirata, H. Tsutsui, N. Tamaki, Cardiac (18)F-FDG PET/CT with heparin detects infective vegetation in a patient with mechanical valve replacement, Clin. Nucl. Med. 37 (12) (2012) 1184–1185. [17] R.J. Gropler, Methodology governing the assessment of myocardial glucose metabolism by positron emission tomography and fluorine 18-labeled fluorodeoxyglucose, J. Nucl. Cardiol. 1 (1994) S4–14. [18] O. Wieben, C. Francois, S.B. Reeder, Cardiac MRI of ischemic heart disease at 3T: potential and challenges, Eur. J. Radiol. 65 (1) (2008) 15–28. [19] M. Saeed, M.F. Wendland, N. Watzinger, H. Akbari, C.B. Higgins, MR contrast media for myocardial viability, microvascular integrity and perfusion, Eur. J. Radiol. 34 (3) (2000) 179–195. [20] A.A. Qayyum, P. Hasbak, H.B. Larsson, T.E. Christensen, A.A. Ghotbi, A.B. Mathiasen, N.G. Vejlstrup, A. Kjaer, J. Kastrup, Quantification of myocardial perfusion using cardiac magnetic resonance imaging correlates significantly to rubidium-82 positron emission tomography in patients with severe
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
coronary artery disease: a preliminary study, Eur. J. Radiol. 83 (7) (2014) 1120–1128. C. Rischpler, N. Langwieser, M. Souvatzoglou, A. Batrice, S. van Marwick, J. Snajberk, T. Ibrahim, K.L. Laugwitz, S.G. Nekolla, M. Schwaiger, PET/MRI early after myocardial infarction: evaluation of viability with late gadolinium enhancement transmurality vs 18F-FDG uptake, Eur. Heart J. Cardiovasc. Imaging 16 (6) (2015) 661–669. S.G. Nekolla, C. Rischpler, Assessing myocardial metabolism with hybrid PET imaging: instrumentation concepts, and workflows, Curr. Pharm. Des. 22 (1) (2015) 96–104. N. Bettencourt, A. Chiribiri, A. Schuster, E. Nagel, Assessment of myocardial ischemia and viability using cardiac magnetic resonance, Curr. Heart Fail. Rep. 6 (3) (2009) 142–153. A. Kino, S. Zuehlsdorff, J.J. Sheehan, P.J. Weale, T.J. Carroll, R. Jerecic, J.C. Carr, Three-dimensional phase-sensitive inversion-recovery turbo FLASH sequence for the evaluation of left ventricular myocardial scar, AJR Am. J. Roentgenol. 193 (5) (2009) W381–388. A. Schultz, T. Caspar, M. Schaeffer, A. Labani, M.Y. Jeung, S. El Ghannudi, C. Roy, M. Ohana, Late gadolinium enhancement cardiac imaging on a 3T scanner with parallel, RF transmission technique: prospective comparison of 3D-PSIR and 3D-IR, Eur. Radiol. 26 (6) (2016) 1547–1555. M. Osterholt, S. Sen, V. Dilsizian, H. Taegtmeyer, Targeted metabolic imaging to improve the management of heart disease, JACC Cardiovasc. Imaging 5 (2) (2012) 214–226. R.J. Gropler, E.M. Geltman, K. Sampathkumaran, J.E. Pérez, S.M. Moerlein, B.E. Sobel, S.R. Bergmann, B.A. Siegel, Functional recovery after coronary revascularization for chronic coronary artery disease is dependent on maintenance of oxidative metabolism, J. Am. Coll. Cardiol. 20 (3) (1992) 569–577. R.J. Gropler, R.S. Beanlands, V. Dilsizian, E.D. Lewandowski, F.S. Villanueva, M.C. Ziadi, Imaging myocardial metabolic remodeling, J. Nucl. Med. 51 (Suppl. 1) (2010) 88S–101S. S. Fürst, R. Grimm, I. Hong, M. Souvatzoglou, M.E. Casey, M. Schwaiger, S.G. Nekolla, S.I. Ziegler, Motion correction strategies for integrated PET/MR, J. Nucl. Med. 56 (2) (2015) 261–269.
9
[30] A. Martinez-Möller, S.G. Nekolla, Attenuation correction for PET/MR: problems, novel approaches and practical solutions, Z. Med. Phys. 22 (4) (2012) 299–310. [31] J.M. Lau, R. Laforest, H. Sotoudeh, X. Nie, S. Sharma, J. McConathy, E. Novak, A. Priatna, R.J. Gropler, P.K. Woodard, Evaluation of attenuation correction in cardiac PET using PET/MR, J. Nucl. Cardiol. (October) (2015) [Epub ahead of print] PubMed PMID: 26499770. [32] G. Heusch, R. Schulz, Myocardial hibernation: adaptation to ischaemia, Eur. Heart J. 17 (6) (1996) 824–828. [33] G. Heusch, K.R. Sipido, Myocardial hibernation: a double-edged sword, Circ. Res. 94 (8) (2004) 1005–1007. [34] R.O. Bonow, G. Maurer, K.L. Lee, T.A. Holly, P.F. Binkley, P. Desvigne-Nickens, J. Drozdz, P.S. Farsky, A.M. Feldman, T. Doenst, R.E. Michler, D.S. Berman, J.C. Nicolau, P.A. Pellikka, K. Wrobel, N. Alotti, F.M. Asch, L.E. Favaloro, L. She, E.J. Velazquez, R.H. Jones, J.A. Panza, STICH Trial investigators myocardial viability and survival in ischemic left ventricular dysfunction, N. Engl. J. Med. 364 (17) (2011) 1617–1625. [35] E.J. Kong, S.H. Lee, I.H. Cho, Myocardial fibrosis in hypertrophic cardiomyopathy demonstrated by integrated cardiac F-18 FDG PET/MR, Nucl. Med. Mol. Imaging 47 (3) (2013) 196–200. [36] K. Ito, M. Morooka, O. Okazaki, R. Minaminoto, K. Kubota, M. Hiroe, Efficacy of heparin loading during an 18F-FDG PET/CT examination to search for cardiac sarcoidosis activity, Clin. Nucl. Med. (2013) 128–130. [37] S. Schneider, A. Batrice, C. Rischpler, M. Eiber, T. Ibrahim, S.G. Nekolla, Utility of multimodal cardiac imaging with PET/MRI in cardiac sarcoidosis: implications for diagnosis, monitoring and treatment, Eur. Heart J. 35 (February (5)) (2014) 312. [38] P. Heusch, C. Buchbender, K. Beiderwellen, F. Nensa, V. Hartung-Knemeyer, T.C. Lauenstein, A. Bockisch, M. Forsting, G. Antoch, T.A. Heusner, Standardized uptake values for [18 F] FDG in normal organ tissues: comparison of whole-body PET/CT and PET/MRI, Eur. J. Radiol. 82 (May (5)) (2013) 870–876.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
ARTICLE IN PRESS European Journal of Radiology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Imaging of the myocardium using 18 F-FDG-PET/MRI Jiˇrí Ferda (M.D. Ph.D Prof.) a,∗ , Milan Hromádka (M.D. Ph.D) b , Jan Baxa (M.D. Ph.D) a a b
Clinic of the Imaging Methods, University Hospital Plzen, Alej Svobody 80, 304 60 Plzen, ˇ Czech Republic Department of Cardiology, University Hospital Plzen, Alej Svobody 80, 304 60 Plzen, ˇ Czech Republic
a r t i c l e
i n f o
Article history: Received 23 March 2016 Received in revised form 17 July 2016 Accepted 19 July 2016 Keywords: Myocardium Positron emission tomography PET/MRI Fluorodeoxyglucose
a b s t r a c t The introduction of the integrated hybrid PET/MRI equipment creates the possibility to perform PET and MRI simultaneously. Depending on the clinical question, the metabolic conversion to glycolytic activity or beta-oxidation is performed before the application of FDG. Since FDG aids to evaluate the energetic metabolism of the myocytes and myocardial MRI reaches the imaging capabilities of perfusion and tissue characterization in the daily routine, FDG-PET/MRI looks to be a promising method of PET/MRI exploitation in cardiac imaging. When myocardial FDG uptake should be evaluated in association with the perfusion distribution, the cross-evaluation of FDG accumulation distribution and perfusion distribution pattern is necessary. The different scenarios may be used in the assessment of myocardium, the conversion to glycolytic activity is used in the imaging of the viable myocardium, but the glycolytic activity suppression might be used in the indications of the identification of injured myocardium by ischemia or inflammation. FDG-PET/MRI might aid to answer the clinical tasks according to the structure, current function and possibilities to improve the function in ischemic heart disease or to display the extent or activity of myocardial inflammation in sarcoidosis. The tight coupling between metabolism, perfusion and contractile function offers an opportunity for the simultaneous assessment of cardiac performance using one imaging modality. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hybrid imaging has brought a powerful tool to clinical practice that combines several advantages of both morphological and metabolic imaging. During the last two decades, 18 F-fluorodeoxyglucose (FDG) has changed the clinical spread of positron emission tomography (PET) imaging like no other molecule. The universal and achievable behaviour of FDG predicts it to be widely available every day and everywhere. Since its first years, PET/CT using FDG has played a crucial role in oncological imaging. Due to the technical difficulties in the co-existence of the PET acquisition system with the high magnetic field, the clinical use of PET/MRI was delayed for about fifteen years. Oncological imaging using FDG also still looks dominant in the current indications of PET/MRI. The introduction of the integrated hybrid PET/MRI equipment creates the possibility to perform PET and MRI simultaneously. The advent of this new technique has introduced the idea of new concepts in myocardial imaging [1,2], respecting all contraindications of the MRI in patients with incompatible implants
∗ Corresponding author. E-mail addresses: [email protected] (J. Ferda), [email protected] (M. Hromádka), [email protected] (J. Baxa).
or claustrophobia. As magnetic resonance is the gold standard in vivo technique in the assessment of the myocardial structure, positron emission tomography has become the gold standard of the in vivo investigation of its metabolism. The synergy in myocardial imaging should be the driving force of myocardial PET/MRI concept. Due to the advantage of the choice, the selection of the most effective way to assess myocardial perfusion, metabolism and function has to be evaluated and the possible role of myocardial PET/MRI has to be claimed. Since FDG aids to evaluate the energetic metabolism of the myocytes [3] and myocardial MRI reaches the imaging capabilities of perfusion and tissue characterization in the daily routine, FDG-PET/MR looks to be a promising method of PET/MRI exploitation in cardiac imaging, even if other tracers like 13 N-ammonium, 15 O-water, 82 Rb-rubidium-chloride, 18 F-flurpiridaz or 11 C-palmitate may be used in cardiac PET.
2. Myocardial energy metabolism and FDG uptake A major role of FDG use in PET myocardial imaging is played by the relationship between -oxidation or oxidative glycolysis based energy metabolism [4]. Myocardial energy metabolism depends on the oxidation of various substrates. Under fasting conditions, the myocardium uses almost all energy for the production of adenosine
http://dx.doi.org/10.1016/j.ejrad.2016.07.014 0720-048X/© 2016 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
2
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
triphosphate (ATP) during the oxidation of free fatty acids [5]. Fasting decreases plasma glucose levels rapidly, and the free fatty acids levels become high. The extraction of fatty acids by the myocardium is enhanced, and long-chain acyl-coenzym A (CoA) is increasingly formed. Even the activated fatty acids could be used in triacylglycerol or phospholipids building, as fasting facilitates their role in energy metabolism. Fatty acid energy-based metabolism takes place inside the mitochondrion. Acyl-CoA is useful as a source of substrate of -oxidation and the majority is used in ATP synthesis. However, the molecule of acyl-CoA is too large to be transported into the mitochondria, the carnitine shuttle aids to pass the longchain fatty acids with more than 15 acetate parts. The product of the -oxidation acetyl-CoA then enters the cycle of Krebs (tricarboxylic acid cycle), the final pathway of the oxidative metabolism of all substrates. When the insulin plasma level is low and free fatty acids levels increase, only a small amount of the glucose is extracted by the myocardium [6,7]. However, when the glucose level increases, the energetic role of the oxidative glycolysis becomes more important. In a postprandial state, one-third to one-half of energy used by the myocardium depends on oxidative glycolysis [8]. Beyond the physiological conditions, the state after myocardial ischemia is the situation in which the up-regulation of the oxidative glycolysis becomes the one-way energetic pathway [9–11]. Nonetheless, not only the acute situation of the ischemia enhances the glycolysis; the glucose metabolism also remains for several weeks after the ischemic episode, partially due to inflammatory cell immigration into the insulted myocardium, partially when the myocardium is hibernating, even if the myocardium is recovering in normal perfusion conditions and functional ability (Fig. 1). The FDG molecule looks very similar to that of glucose, only the OH group is replaced by the fluorine. A very similar molecular appearance is responsible for FDG behaviour [3]. FDG accumulates in the myocardial tissue proportionally according to the glucose transport and phosphorylation, and even FDG exhibits a slightly different affinity to transporters and phosphatase compared to glucose. The utilization of the glucose does not have a uniform pattern in the entire heart, and, consequently, the inhomogeneous FDG uptake distribution is obvious [8]. Copying the glucose metabolism, the lateral wall of the left ventricle exhibits a more homogenous and more intense distribution of FDG [9,12]. Because of the outlined variability of the energetic sources, the major role in myocardial metabolic conversion belongs to the patient’s dietary preparation and pharmacological intervention. As during fasting conditions, the glucose levels are decreasing, therefore the metabolism of the myocardium converts towards fatty acid -oxidation. This fact makes the PET images after intravenous administration of FDG poor and of unstable quality [13]. However, the fasting state before the imaging has been applied to detect the ischemic myocardium after physical exercise. Fasting helps to identify myocardial infarction, infarcted tissue exhibits an increased level of glucose metabolism not only due to the myocardial glucose conversion, but also due to the migration of inflammatory vectors such as macrophages and/or activated fibroblasts. To be able to delineate the normal myocardium, it is necessary to make its glucose metabolism as low as possible. An effective way to suppress glycolytic activity in normal myocytes is a low carbohydrate and high fat diet (the Atkins diet) [14]. There is also an alternative way to myocardial conversion to fatty acid exploitation. The intravenous administration of un-fractioned heparin leads to rapid conversion to -oxidation in the myocardium. Having a direct impact on lipoprotein lipase activation, the heparin causes the elevation of the plasma free fatty acids levels and switches the normal myocytes almost exclusively to -oxidation [15]. The metabolic switch almost exclusively to glucose is the important challenge in the assessment of myocardial viability. The main strategy is to offer to the myocardium such a glucose overload
that myocytes are able to diminish the -oxidation and the forced glucose extraction results in high FDG uptake. There is a traditional approach and an alternative one. Some centres advocate the hyperinsulinemic-euglycemic clamp – to intravenously administer insulin and glucose simultaneously under control of normal glucose plasma levels. The main intent is to standardize the glucose metabolism within the whole myocardium, but the most important disadvantage is the complicated procedure [16]. An easier way to convert the myocardium to glycolysis is the oral glucose load. Oral glucose administration increases plasma insulin level and accelerates the glycolytic activity and decreases the plasma fatty acid level. Oral glucose load very effectively forces the conversion in non-diabetic patients. In diabetics, there are two main problems in myocardium glycolytic conversion. High glucose plasma levels lead to the limited ability of the myocardium to extract FDG. The second very important problem could be the insensitivity of some diabetics to insulin, which has a key role in metabolic switching [6,7]. The classic approach of the FDG uptake evaluation is based on proportional assessments according to myocardial perfusion [16,17]. Proven concepts include the perfusion evaluation using PET tracers such as 15 O-H2 O, 13 N-NH3 or 82 Ru. PET-based perfusion imaging could be successfully replaced by SPECT imaging using 99m Tc labelled methoxy-lisobuthyl-isonitril or 201 Tl. However, cardiac magnetic resonance offers perfusion imaging based on the first pass of the gadolinium contrast agent (Fig. 4).
3. Myocardial gadolinium imaging during PET/MRI Gadolinium-based contrast agents have been used for thirty years to involve the tissue behaviour in T1 weighted images (with a known influence also in T2 weighted images). Gadolinium based contrast agents (GBCA) have also been used in myocardial imaging for the past two decades, and the possibilities of myocardial imaging using magnetic resonance and the GBCA application have been studied intensively. Concerning the myocardium, multiparametric ECG synchronized cardiac magnetic resonance imaging (CMRI) included kinetic studies, the first past perfusion imaging and the evaluation of the myocardial structure, including the late gadolinium enhancement evaluation. Using the basic sequences, the associate findings such as pericardial fluid or valvular disease could be found at the same time. With respect to specific questions like flow quantification, valve regurgitation or contractile myocardial function, the sequences targeting blood flow (phase contrast imaging) or myocardial contractility (myocardial tagging) may be used. Multiple studies assessed the clinical relevance of flow-limiting coronary artery disease (CAD). CMRI detected more cases of obstructive coronary artery disease without increasing the falsepositive rate; however, the sensitivity of nuclear stress testing is significantly lower than had previously been published. Very few centres now use CMRI for the diagnosis of CAD given the lack of experienced clinicians able to perform the study, the inability to perform exercise stress CMRI, and patient issues including claustrophobia and relatively long imaging times. Nevertheless, studies were able to show that CMRI can serve as an alternative to stress echocardiography and nuclear stress testing when evaluating patients for obstructive CAD (Fig. 7). When myocardial FDG uptake should be evaluated in association with the perfusion distribution, the cross-evaluation of FDG accumulation distribution and perfusion distribution pattern is necessary. Modelling of the myocardial flow using GBCA is completely different than that used in PET. PET flow modelling is based on indicator extraction by cardiomyocytes, so it uses intracellular uptake of the tracer. Because neither the time resolution nor
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
3
Fig. 1. A male patient after acute myocardial infarction and consequent angioplasty of the left anterior descending artery, images in short axis view. Multiparametric assessment of the myocardium: (a) – early phase of the first pass perfusion using turboFLASH T1 sequence, (b) – late phase of the first pass perfusion using turboFLASH T1 sequence, (c) – late gadolinium enhancement image using phase sensitive inversion recovery sequence, (d) – perfusion analysis – maximum slope map representing the perfusion flow function, (e) – perfusion analysis – initial area under the curve map representing the flow volume, (f) – FDG distribution on PET image representing the level of glycolytic activity. Within the lateral wall of the left ventricle, there is the hypo-perfusion zone with sub-endocardial infarction exhibiting the sub-endocardial late enhancement, the metabolism is diffusely decreased in the same region.
Fig. 2. A male patient one year after large myocardial infarction in the interventricular septum suffering from low left ventricle ejection fraction of 20%, images in short axis view, patient prepared with oral glucose load, PET/MR fusion: (a) – pure FDG-PET image, (b) fused FDG-PET/MR image, (c) – late gadolinium enhancement MR image. The infarction of the septum contains the islets of the tissue without LGE, but accumulating FDG. The tissue exhibits the viable population of the myocytes.
the spatial resolution is sufficient, no first pass imaging is possible. Magnetic resonance imaging is traditionally based on first pass perfusion; it exploits the T1 weighted steady state sequences such as turboFLASH. Other MRI related perfusion techniques such as gadolinium-based T2* perfusion, arterial spin labelling (ASL) and BOLD (blood oxygen level dependent) imaging have not been widely distributed. When three Tesla systems enable high resolution, the steady state T1 weighted images might be used by hybrid integrated PET/MRI equipment [18,19]. Depending on the heart rate of an individual, three to six planes could cover the myocardium. The first pass of the contrast agent through the coronary riverbed takes about one breath hold, thus the perfusion data set could be acquired within one sustained breath hold (Fig. 10).
GBCA perfusion imaging tracks the advent of the indicator into the micro-vascular extracellular space of the myocardium and then the evaluation of the saturation of the myocardium during one perfusion cycle [20,21]. When the signal of the myocardium depends on the own properties of the used acquisition sequence, the longitudinal relaxation of the myocardium (related to the water and protein content) and the behavioural properties of the contrast agents (their concentration and/or relaxivity), the quantitative computation of the perfusion within the myocardium might be difficult, even without respect to the movements of the myocardium. Nevertheless, the FDG uptake quantification and standardization within the myocardium is disputable, too. The quantification of the perfusion depends on several factors: the fashion of the bolus appli-
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
4
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
Fig. 3. The same patient like in Fig. 2, MR first pass perfusion: (a) – early phase, (b) mid-phase, (c) – later phase. The septum contains the tissue with preserved perfusion, corresponding to that showed by FDG uptake.
Fig. 4. The same patient like in Figs. 2 and 3, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – anterior wall shows preserved normal perfusion (red), lateral wall with moderate decrease of flow (yellow), inferior wall (green) and septum (blue) suffer from decreased perfusion. Comparison of the perfusion and metabolic patterns enables to distinguish between hibernating myocardium (inferior wall) and scar (septum) with the additional information about islets of viable tissue within the septum. The coronary bypass graft surgery improved the perfusion conditions so, that the left ventricle ejection fraction increased to 35%.
Fig. 5. A male patient five days after large myocardial infarction after occlusion of the left anterior descending artery with akinetic apical third of the left ventricle, images in short axis view, patient prepared with intravenous administration of heparin, PET/MR fusion: (a) – pure FDG-PET image, (b) fused FDG-PET/MR image, (c) – late gadolinium enhancement MR image. The infarction exhibits subendocardial localization with late gadolinium enhancement, present also intraventricular thrombus, the tissue which has injured by the ischemia exhibits extreme FDG uptake, it represents “ischemic memory” of the myocardium.
cation, the concentration and the relaxivity of the used agent, and last but not least the computation model [22]. 15 O-water has to be established as the gold standard perfusion indicator with the linear relation of the extraction independent of the blood flow. All other radiotracers suffer from decreasing extraction in higher flow rates in the progressive row 18 F-flurpiridaz, 13 N-ammonium, 82 Rb-rubidium-chloride [22]. Some opinions anticipate the incomparable perfusion assessment between gadolinium first pass and nuclear medicine methods, but the results of the larger studies and especially results of meta-analyses have confirmed the
clinical relevant results of CMRI perfusion quantification. The perfusion quantification computation model presents equal results independently to the algorithm, even if the model is linear (Fermiconstrained deconvolution), model-independent deconvolution, or one-compartment model or an uptake model (similar to the models of radiopharmaceutical extraction) [22]. The most important advantage of the GBCA is that fact that it acts twice. First pass perfusion is followed by the period of late gadolinium enhancement (LGE) after about ten minutes. The LGE phenomenon is based on two different mechanisms. First, the GBCA
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
5
Fig. 6. The same patient like in Fig. 5, MR first pass perfusion: (a) – early phase, (b) mid-phase, (c) – later phase. The wall contains the tissue with preserved perfusion, except the thin inner rim of subendocardial infarction.
Fig. 7. The same patient like in Figs. 5 and 6, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – non infarcted wall shows preserved normal perfusion (red), zone of subendocardial infarction shows decreased flow (green), the tissue with FDG uptake represents stunned myocardium, the contractility recovered after revascularization.
Fig. 8. A female patient with multiorgan involvement by the sarcoidosis, patient prepared with low carbohydrate high fat diet, PET/MR fusion in horizontal axis: (a) – late gadolinium enhancement MR image, (b) fused FDG-PET/MR image, the FDG uptake is copying the presence of the late gadolinium enhancement and showing the sites of the active inflammation, the maximum activity showed the lateral wall.
kinetics are much slower, and second, the volume of distribution is much higher in the infarcted myocardium. LGE is visible due to the different distribution and release of the gadolinium from the interstitial tissue in the conditions of normal myocardium, myocardium affected by the ischemic necrosis or scar tissue [23]. Nevertheless, the indicator delayed kinetics is the basic principle aids to distinguish between myocardium having its integrity from that which is deteriorated by the inflammation and/or fibrosis dur-
ing acute myocarditis or granulomatous inflammation [23,24]. LGE imaging is traditionally based on approach to use the inversion recovery technique to nullify the signal of the normal myocardium and to make visible gadolinium enhancement, typically using 3D-IR-turboFLASH sequences, or to use the phase sensitive inversion recovery (PSIR). The reconstruction of the phase sensitive images improves splitting the signal of the normal myocardium and gadolinium to the opposite points [24]. On three Tesla systems,
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
6
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
Fig. 9. The same patient like in Fig. 8, PET/MR fusion in horizontal axis: (a) – late gadolinium enhancement MR image, (b) fused FDG-PET/MR image, the FDG uptake is copying the presence of the late gadolinium enhancement and showing the sites of the active inflammation, note the subendocardial involvement of the septum, but also the extreme finding of the interlobular fibrosis within the liver affected by the sarcoidosis too.
Fig. 10. The same patient like in Figs. 8 and 9, MR first pass perfusion analysis: (a) – maximum slope map representing the perfusion flow function, (b) – initial area under the curve map representing the flow volume, (c) – time – intensity evaluation curves – the perfusion within the FDG accumulating tissue affected by the active sarcoidosis is decreased similarly like in those conditions seen in myocardial ischemia cases.
the contrast-to-noise ratio (CNR) could be improved without the loss of diagnostic accuracy when PSIR is used [25]. The PSIR singleshot technique allows for imaging a set of slices covering the entire myocardium during a single breath hold without adaptation of the inversion time. 4. Myocardial FDG-PET/MRI acquisition protocol and data evaluation Depending on the clinical question, the metabolic conversion to glycolytic activity or beta-oxidation is performed before the application of FDG. Due to long-lasting data acquisition enabling the use of a sufficient number of coincident events, the FDG dosage could drop to 2.5 MBq/kg of body weight. The optimal accumulation time ranges between 45 and 90 min after the actual beginning of FDG intravenous administration [16,25–27]. Since the heart is a continually moving structure, proper data acquisition is crucial for obtaining valuable results. Both PET and MRI acquisition is driven by the electrocardiogram, the prospective triggering is used in PET data acquisition, and alternating prospective acquisition triggering and retrospective data segmentation are used in the MRI part. Due to breathing, breath-navigated acquisition is combined with ECG in the dual-triggering method. Because of the fundamental importance of the attenuation correction (AC), the two point Dixon T1 spoiled gradient echo sequences (VIBE – volume interpolated breath hold examination) followed immediately after the planning sequences are completed [27–29]. With the start of the AC sequence, the PET acquisition is automatically initiated, and the time of the PET data acquisition
might be selected according to the scanning time of the whole CMRI protocol. In between PET data acquisition of kinetic sequences occur in four chamber, longitudinal two chamber and left outflow two chamber views using steady state free precession sequences (trueFISP) or ultrafast gradient echo sequences (turboFLASH). As an option, myocardial contractility may be assessed in which the tagged gradient echo images are performed before the application of the dynamic post-contrast T1 imaging. Myocardial first pass perfusion sequences are based on T1 weighted turboFLASH sequences repeated through the cardiac cycle during the application of the contrast material bolus. Dosed by 0.5 mmol/kg, preferable are highconcentration (gadobutrol) or high-relaxivity agents (gadobenate dimeglumine) providing the rapid slope of the blood signal. Waiting for the late gadolinium enhancement for ten minutes, the short axis sequences covering the whole heart are scanned. LGE images in diastolic phase complete the protocol, and IR-turboFLASH or PSIR sequences are useful as outlined above. The whole protocol lasted 20 min. The sequences with parameters are included in the table. When data acquisition is completed, the PET reconstruction is started. The tissue model (-maps) based on fat and water image derived from opposed-phase and in-phase is created for use in attenuation corrected image reconstruction [27–29]. Possible problems could occur during the presence of the metal implants resulting in correction failure thanks to signal voids on -maps. Nevertheless, these problems don’t often occur even if cerclage of the sternum is present. The algorithm of the ordered subset with maximization expectation (OSEM) in zoomed field of view might be recommended. If the point spread function is used, the increased noise could be achieved, but use of the reconstruction algorithm
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
improve the spatial signal resolution. In order to be comparable with LGE MRI images, the AC PET images are reconstructed in the diastolic phase. The data evaluation is based on a multi-parametric view of the dedicated software platform. The cardiac perfusion might be assessed using the computation of the perfusion maps related to the 17-segment model recommended by the American Heart Association (AHA). The slice-to-slice fusion reordering of the AC PET images with the morphological MRI images helps compare the level of FDG uptake with the structural changes of the myocardium on CMRI – thickness, contractility, hypo-perfusion and/or delayed enhancement. The calculation of the cardiac function including left ventricle ejection fraction, stroke volume and cardiac output are needed to complete the entire information about the myocardium. 5. Current reasonable FDG-PET/MRI imaging strategies 5.1. Viable myocardium Being the classical indication of both worlds, imaging of the viable myocardium may become the crucial indication of hybrid PET/MRI with FDG, even if both methods are overlapping in some aspects of the imaging. When the clinical question is to estimate the possible improvement of the ventricular function after planned revascularization, the detection of the hibernating myocardium is the target of the imaging. The hibernating myocardium survives in conditions of saturation by oxygen and energy substrates on the edge between a sustainable level of basal metabolism and ischemia leading to necrosis. Usually the flow limiting central stenosis causes hypoperfusion and it tears the myocytes apart from sufficient supply by the blood stream, and consequently the myocardium is left in chronic hypoxia. As outlined above, the hypoxia is the driving force in myocardial conversion to glycolytic metabolism. In order to prevent the problems in heterogeneity of energy substrates exploitation, the glycolytic conversion of the normal myocardium is also needed. For the purpose of providing the viable myocardium the highest level of FDG uptake, the patient has to be prepared with fasting with subsequent glucose load, either by oral load or using the hyperinsulinemic-euglycemic clamp. By glucose driven myocardial metabolism might warrant that no viable myocytes population remained undetected. The conventional approach in the evaluation of FDG accumulation within the myocardium demands comparison with the perfusion pattern. As for providing the useful information, first pass gadolinium MR perfusion completes the information. When PET imaging suffers from insufficient spatial resolution to distinguish between thinned myocardium and partially infarcted as in the sub-endocardial layer, LGE MRI supplements this disadvantage with its excellent ability to discriminate scar regions from those that are structurally normal. FDG-PET/MRI might help in fulfilling all the clinical tasks according to the structure, current function and possibilities to improve the function. 5.2. Myocardial ischemia and post-infarction imaging Complete deregulation of the glycolytic energy metabolism in hypoxic myocardium is the leading idea in the direct imaging of the hypoxic or ischemic myocardium [30,31]. When occurring, suddenly myocytes are turned to the pure glycolysis, no available oxygen makes the anaerobic glycolysis the leading way of the energy metabolism. The conversion occurs rapidly in several minutes (even when induced by stress testing). When ischemia occurs, glycolysis remains and lasts after the hypoxic/ischemic period for several days to weeks, even if the blood supply is restored. The up-regulation of glycolysis by the myocytes is one mecha-
7
nism of the increased FDG accumulation within the post-ischemic myocardium; on the other hand, the ischemic necrosis stimulates the migration of the cellular vectors into the affected tissue. Like induced inflammation with monocytes, activated lymphocytes, activated fibroblasts and progenitor cells are a major cause of the increased glucose metabolism inside the infarcted zone forming scar tissue. These inflammatory cells exhibit extremely high levels of glucose transporters and hexokinase activity. In such a case, thorough glycolysis suppression has to be performed. Intravenous un-fractioned heparin administration is the best way to complete metabolic conversion. The dose of 50 IU of heparin per kilogram of body weight may be recommended. Having some indication problems in subacute myocardial infarction, the Atkins diet appears more problematic to use. The evaluation relies on a direct comparison with first pass MRI perfusion and LGE. The gadolinium late stain shows zones of inflammatory tissue response and scar forming, the restored perfusion without late enhancement represents those zones remembering the ischemia. Although stress-induced hypoxic myocardium imaging is not yet clinically established, the infarction-related changes and ischemic memory of the myocardium could be routinely used [32]. Owing to the unsuspected problems in preparation (fasting, glucose overload or heparin application), both with limited experience with myocardial behaviour in stress related to glucose extraction in humans, the further testing of myocardium imaging has to be performed and critically compared to other stress testing methods [18,29,33,34]. Despite the first reports in the assessment of hypertrophic cardiomyopathy [35], there are no clear advantages of PET/MRI versus MRI. 5.3. Myocardial inflammation The inflammatory cellular population reaches one of the highest grades of glucose consumption according to the up-regulation described in the previous paragraph in cases of post-infarction inflammatory response. In its own sense, myocardial inflammatory diseases represent the migration of the activated cells – either monocytes, activated lymphocytes or neutrophils in infectious or para-infectious acute myocarditis, or histiocytes-related elements, such as epitheloid or Langerhans polynuclear cells in granulomatous diseases represented by sarcoidosis [36,37]. The acute onset of myocardial infiltration is linked to the elevation of myocardial biomarkers, especially troponin. Myocardial injury not related to ischemia is one of the most frequent indications of CMRI. To achieve the optimum effectiveness of the discrimination between nonaffected myocardium, active inflammation and non-active fibrous scar tissue, the conversion to -oxidation is needed [16,36,37]. Glycolysis suppression in healthy myocardium and late gadolinium images makes the comparison of the metabolism and structure possible. Having late gadolinium enhancement, the FDG avid tissue represents active inflammation. On the other hand, low accumulation confirms non-active fibrosis in the zone of late gadolinium enhancement. The normal myocardium remains non-enhanced and non-accumulating FDG. 6. Future perspectives of myocardial imaging using FDG-PET/MRI The future wider acceptance of the FDG-PET/MRI as an imaging tool for myoacardial assessment is depending on the future distribution of the equipment. For the future development of indications is very important advantage radiation dose burden to the patient during FDG-PET/MRI. The use of 2.5 MBq/kg in 70 kg patient is related to the dose about 3,5 mSV, that dose is similar to the abdominal CT examination.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS
8
J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
When the myocardial injury related or not related to ischemia is still increasing indication of CMRI, the imaging of the activity of the inflammation could plays a role in the prognosis estimation in patients with wide range of causes of myocardial impairment – acute ischemia, myocarditis or sarcoidosis. More precise detection of the process extent and its activity using a combination of the T1 and T2 mapping, water molecule diffusion and perfusion imaging might with imaging of the metabolically active nonmyocyte-cellular population. In the near future, that combination of multiparametric imaging of the metabolism, microstructure and microvascular perfusion could aid to achieve the optimum effectiveness of the discrimination between non-affected myocardium, active inflammation and non-active fibrous tissue. Simple and quick conversion to -oxidation using non-fractioned heparin makes the direct inflammation imaging easier, so the increased role of the metabolic suppression of the healthy myocardium could be awaited. Even the original indication of FDG-PET was the imaging of the viable myocardium, the future use will be further in the shadow of LGE. And what about the imaging beyond fluorodehydroxyglucose? Besides the FDG imaging, the other radiopharmaceuticals are able to play some role in myocardial imaging. The hypoxia imaging has some potential in the estimation of the hibernating myocyte population, the promising should be fluoromisonidazol of 64 Cu labelled substances. The special requests for amyloid imaging within myocardium maybe fulfil the drugs similar like florbetaben or flutemetamol, whose are used for brain amyloid imaging. The investigation of the neural loss in some patients could be performed using fluorodihydroxyphenylalanine. But the spread of these methods will last more time, because of needed effectivity and safety studies. For a closer future, the myocardial PET/MRI will still remain the FDG based.
7. Conclusion As FDG is a common radiopharmaceutical used in PET imaging and as there is long experience with CMRI, the natural combination of both methods in PET/MRI became the first choice in the introduction into clinical practice. Metabolic imaging combined with excellent structural information, even more supplemented with perfusion imaging using MRI is one of the major strengths. Also there are no problems with short-living isotopes needing on-site cyclotrons. While the sophisticated viability assessment looks like the easiest way to clinical adoption, the hypoxic/ischemic memory or infarction extend assessments need clinical relevant questions [18,19]. The final advantage, the lower radiation dose than is used in one chest CT, makes PET/MRI preferable to PET/CT, when the results of PET remain comparable [38]. However, while myocardial FDG-PET/MRI is promising, the other kinds of radiopharmaceuticals remain in the shadows. Other radiopharmaceuticals are widely useful, some are still not registered (18 F-flurpiridaz in perfusion imaging), registered in other indication (18 F-florbetaben in amyloid imaging, 18 F-fluorodihydroxyphenyalanine in dopaminergic imaging), closely related to cyclotron availability (13 N-ammonium, 15 O-water or 11 C-palmitate), or bearing the high radiation dose burden to the patient (82 Rb-rubidium-chloride). In the coming years, further evolution of the method might be observed.
Conflict of interest The authors have no conflict of interest to disclose.
Authors contribution We wish to confirm there has been no significant financial support for this work that could have influenced its outcome. All authors have read and approved submission of this manuscript. This work has not been published and is not being considered for publication elsewhere. Acknowledgements Supported by project Conceptual development of the research institution of the Czech Ministery of Health No. 00669806 – FN ˇ and by Program of the development of the Charles´ı University Plzen (project P36). References [1] C. Rischpler, S.G. Nekolla, K.P. Kunze, M. Schwaiger, PET/MRI of the heart, Semin. Nucl. Med. 45 (3) (2015) 234–247. [2] C. Rischpler, S.G. Nekolla, I. Dregely, M. Schwaiger, Hybrid PET/MR imaging of the heart: potential, initial experiences, and future prospects, J. Nucl. Med. 54 (3) (2013) 402–415. [3] A.J. Liedtke, B. Renstrom, S.H. Nellis, Correlation between [5-3H]glucose and [U-14C] deoxyglucose as markers of glycolysis in reperfused myocardium, Circ. Res. 71 (3) (1992) 689–700. [4] A.J. Liedtke, Alterations of carbohydrate and lipid metabolism in the acutely ischemic heart, Prog. Cardiovasc. Dis. 23 (5) (1981) 321–336. [5] P.1 Camici, E. Ferrannini, L.H. Opie, Myocardial metabolism in ischemic heart disease: basic principles and application to imaging by positron emission tomography, Prog. Cardiovasc. Dis. 32 (3) (1999) 217–238. [6] D.1 Sun, N. Nguyen, T.R. DeGrado, M. Schwaiger, FC 3rd. Brosius, Ischemia induces translocation of the insulin-responsive glucose transporter GLUT4 to the plasma membrane of cardiac myocytes, Circulation 89 (2) (1994) 793–798. [7] S.1 Egert, N. Nguyen, M. Schwaiger, Myocardial glucose transporter GLUT1: translocation induced by insulin and ischemia, J. Mol. Cell. Cardiol. 31 (7) (1999) 1337–1344. [8] R.C. Marshall, S.C. Huang, W.W. Nash, M.E. Phelps, Assessment of the [18F]fluorodeoxyglucose kinetic model in calculations of myocardial glucose metabolism during ischemia, J. Nucl. Med. 24 (11) (1983) 1060–1064. [9] G. Heusch, The regional myocardial flow-function relationship: a framework for an understanding of acute ischemia, hibernation, stunning and coronary microembolization. 1980, Circ. Res. 112 (12) (2013) 1535–1537. [10] G. Heusch, R. Schulz, S.H. Rahimtoola, Myocardial hibernation: a delicate balance, Am. J. Physiol. Heart Circ. Physiol. 288 (3) (2005) H984–999. [11] L.H. Opie, Effects of regional ischemia on metabolism of glucose and fatty acids Relative rates of aerobic and anaerobic energy production during myocardial infarction and comparison with effects of anoxia, Circ. Res. 38 (5 Suppl. 1) (1976) 152–174. [12] A.J. Liedtke, H.C. Hughes, J.R. Neely, Effects of excess glucose and insulin on glycolytic metabolism during experimental myocardial ischemia, Am. J. Cardiol. 38 (1) (1976) 17–27. [13] R.J. Gropler, B.A. Siegel, K.J. Lee, S.M. Moerlein, D.J. Perry, S.R. Bergmann, E.M. Geltman, Nonuniformity in myocardial accumulation of fluorine-18-fluorodeoxyglucose in normal fasted humans, J. Nucl. Med. 31 (11) (1990) 1749–1756. [14] Y. Kobayashi, S. Kumita, Y. Fukushima, K. Ishihara, M. Suda, M. Sakurai, Significant suppression of myocardial (18)F-fluorodeoxyglucose uptake using 24-h carbohydrate restriction and a low-carbohydrate, high-fat diet, J. Cardiol. 62 (5) (2013) 314–319. [15] A.M. Scholtens, H.J. Verberne, R.P. Budde, M. Lam, Additional heparin pre-administration improves cardiac glucose metabolism suppression over low carbohydrate diet alone in 18F-FDG-PET imaging, J. Nucl. Med. (December) (2015), pii:jnumed.115.166884. [Epub ahead of print] PubMed PMID: 26659348. [16] A.1 Klaipetch, O. Manabe, N. Oyama-Manabe, S. Chiba, M. Naya, S. Yamada, K. Hirata, H. Tsutsui, N. Tamaki, Cardiac (18)F-FDG PET/CT with heparin detects infective vegetation in a patient with mechanical valve replacement, Clin. Nucl. Med. 37 (12) (2012) 1184–1185. [17] R.J. Gropler, Methodology governing the assessment of myocardial glucose metabolism by positron emission tomography and fluorine 18-labeled fluorodeoxyglucose, J. Nucl. Cardiol. 1 (1994) S4–14. [18] O. Wieben, C. Francois, S.B. Reeder, Cardiac MRI of ischemic heart disease at 3T: potential and challenges, Eur. J. Radiol. 65 (1) (2008) 15–28. [19] M. Saeed, M.F. Wendland, N. Watzinger, H. Akbari, C.B. Higgins, MR contrast media for myocardial viability, microvascular integrity and perfusion, Eur. J. Radiol. 34 (3) (2000) 179–195. [20] A.A. Qayyum, P. Hasbak, H.B. Larsson, T.E. Christensen, A.A. Ghotbi, A.B. Mathiasen, N.G. Vejlstrup, A. Kjaer, J. Kastrup, Quantification of myocardial perfusion using cardiac magnetic resonance imaging correlates significantly to rubidium-82 positron emission tomography in patients with severe
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),
G Model EURR-7524; No. of Pages 9
ARTICLE IN PRESS J. Ferda et al. / European Journal of Radiology xxx (2016) xxx–xxx
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
coronary artery disease: a preliminary study, Eur. J. Radiol. 83 (7) (2014) 1120–1128. C. Rischpler, N. Langwieser, M. Souvatzoglou, A. Batrice, S. van Marwick, J. Snajberk, T. Ibrahim, K.L. Laugwitz, S.G. Nekolla, M. Schwaiger, PET/MRI early after myocardial infarction: evaluation of viability with late gadolinium enhancement transmurality vs 18F-FDG uptake, Eur. Heart J. Cardiovasc. Imaging 16 (6) (2015) 661–669. S.G. Nekolla, C. Rischpler, Assessing myocardial metabolism with hybrid PET imaging: instrumentation concepts, and workflows, Curr. Pharm. Des. 22 (1) (2015) 96–104. N. Bettencourt, A. Chiribiri, A. Schuster, E. Nagel, Assessment of myocardial ischemia and viability using cardiac magnetic resonance, Curr. Heart Fail. Rep. 6 (3) (2009) 142–153. A. Kino, S. Zuehlsdorff, J.J. Sheehan, P.J. Weale, T.J. Carroll, R. Jerecic, J.C. Carr, Three-dimensional phase-sensitive inversion-recovery turbo FLASH sequence for the evaluation of left ventricular myocardial scar, AJR Am. J. Roentgenol. 193 (5) (2009) W381–388. A. Schultz, T. Caspar, M. Schaeffer, A. Labani, M.Y. Jeung, S. El Ghannudi, C. Roy, M. Ohana, Late gadolinium enhancement cardiac imaging on a 3T scanner with parallel, RF transmission technique: prospective comparison of 3D-PSIR and 3D-IR, Eur. Radiol. 26 (6) (2016) 1547–1555. M. Osterholt, S. Sen, V. Dilsizian, H. Taegtmeyer, Targeted metabolic imaging to improve the management of heart disease, JACC Cardiovasc. Imaging 5 (2) (2012) 214–226. R.J. Gropler, E.M. Geltman, K. Sampathkumaran, J.E. Pérez, S.M. Moerlein, B.E. Sobel, S.R. Bergmann, B.A. Siegel, Functional recovery after coronary revascularization for chronic coronary artery disease is dependent on maintenance of oxidative metabolism, J. Am. Coll. Cardiol. 20 (3) (1992) 569–577. R.J. Gropler, R.S. Beanlands, V. Dilsizian, E.D. Lewandowski, F.S. Villanueva, M.C. Ziadi, Imaging myocardial metabolic remodeling, J. Nucl. Med. 51 (Suppl. 1) (2010) 88S–101S. S. Fürst, R. Grimm, I. Hong, M. Souvatzoglou, M.E. Casey, M. Schwaiger, S.G. Nekolla, S.I. Ziegler, Motion correction strategies for integrated PET/MR, J. Nucl. Med. 56 (2) (2015) 261–269.
9
[30] A. Martinez-Möller, S.G. Nekolla, Attenuation correction for PET/MR: problems, novel approaches and practical solutions, Z. Med. Phys. 22 (4) (2012) 299–310. [31] J.M. Lau, R. Laforest, H. Sotoudeh, X. Nie, S. Sharma, J. McConathy, E. Novak, A. Priatna, R.J. Gropler, P.K. Woodard, Evaluation of attenuation correction in cardiac PET using PET/MR, J. Nucl. Cardiol. (October) (2015) [Epub ahead of print] PubMed PMID: 26499770. [32] G. Heusch, R. Schulz, Myocardial hibernation: adaptation to ischaemia, Eur. Heart J. 17 (6) (1996) 824–828. [33] G. Heusch, K.R. Sipido, Myocardial hibernation: a double-edged sword, Circ. Res. 94 (8) (2004) 1005–1007. [34] R.O. Bonow, G. Maurer, K.L. Lee, T.A. Holly, P.F. Binkley, P. Desvigne-Nickens, J. Drozdz, P.S. Farsky, A.M. Feldman, T. Doenst, R.E. Michler, D.S. Berman, J.C. Nicolau, P.A. Pellikka, K. Wrobel, N. Alotti, F.M. Asch, L.E. Favaloro, L. She, E.J. Velazquez, R.H. Jones, J.A. Panza, STICH Trial investigators myocardial viability and survival in ischemic left ventricular dysfunction, N. Engl. J. Med. 364 (17) (2011) 1617–1625. [35] E.J. Kong, S.H. Lee, I.H. Cho, Myocardial fibrosis in hypertrophic cardiomyopathy demonstrated by integrated cardiac F-18 FDG PET/MR, Nucl. Med. Mol. Imaging 47 (3) (2013) 196–200. [36] K. Ito, M. Morooka, O. Okazaki, R. Minaminoto, K. Kubota, M. Hiroe, Efficacy of heparin loading during an 18F-FDG PET/CT examination to search for cardiac sarcoidosis activity, Clin. Nucl. Med. (2013) 128–130. [37] S. Schneider, A. Batrice, C. Rischpler, M. Eiber, T. Ibrahim, S.G. Nekolla, Utility of multimodal cardiac imaging with PET/MRI in cardiac sarcoidosis: implications for diagnosis, monitoring and treatment, Eur. Heart J. 35 (February (5)) (2014) 312. [38] P. Heusch, C. Buchbender, K. Beiderwellen, F. Nensa, V. Hartung-Knemeyer, T.C. Lauenstein, A. Bockisch, M. Forsting, G. Antoch, T.A. Heusner, Standardized uptake values for [18 F] FDG in normal organ tissues: comparison of whole-body PET/CT and PET/MRI, Eur. J. Radiol. 82 (May (5)) (2013) 870–876.
Please cite this article in press as: J. Ferda, et al., Imaging of the myocardium using http://dx.doi.org/10.1016/j.ejrad.2016.07.014
18 F-FDG-PET/MRI,
Eur J Radiol (2016),