18F-Choline Images Murine Atherosclerotic Plaques Ex Vivo
Objective— Current imaging modalities of atherosclerosis mainly visualize plaque morphology. Valuable insight into plaque biology was achieved by visualizing enhanced metabolism in plaque-derived macrophages using 18F-fluorodeoxyglucose (18F-FDG). Similarly, enhanced uptake of 18F-fluorocholine (18F-FCH) was associated with macrophages surrounding an abscess. As macrophages are important determinants of plaque vulnerability, we tested 18F-FCH for plaque imaging.
Methods and Results— We injected 18F-FCH (n=5) or 18F-FDG (n=5) intravenously into atherosclerotic apolipoprotein E-deficient mice. En face measurements of aortae isolated 20 minutes after 18F-FCH injections demonstrated an excellent correlation between fat stainings and autoradiographies (r=0.842, P<0.0001), achieving a sensitivity of 84% to detect plaques by 18F-FCH. In contrast, radiotracer uptake 20 minutes after 18F-FDG injections correlated less with en face fat stainings (r=0.261, P<0.05), reaching a sensitivity of 64%. Histological analyses of cross-sections 20 minutes after coinjections of 18F-FCH and 14C-FDG (n=3) showed that 18F-FCH uptake correlated better with fat staining (r=0.740, P<0.0001) and macrophage-positive areas (r=0.740, P<0.0001) than 14C-FDG (fat: r=0.236, P=0.29 and CD68 staining: r=0.352, P=0.11), respectively.
Conclusions— 18F-FCH identifies murine plaques better than 18F-FDG using ex vivo imaging. Enhanced 18F-FCH uptake into macrophages may render this tracer a promising candidate for imaging plaques in patients.
Current clinical tools to image atherosclerotic plaques such as intravascular ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), or optical coherence tomography (OCT) provide morphological images of the vessel wall at remarkable spatial resolution.1–4 However, these imaging modalities are unable to characterize details of plaque biology. More insight into plaque biology would be desirable to determine a better risk assessment of the vulnerable plaque.5 The vulnerable plaque, because of its risk of rupturing, is considered the culprit lesion causing acute arterial obstruction, which may result in myocardial infarction or stroke.
Therefore, plaque imaging using molecular targets has gained increased attention.6 For example, positron emission tomography (PET) provides attractive opportunities to visualize plaque biology.7 Specifically, 18F-fluorodeoxyglucose (18F-FDG)—a widely used PET tracer—has been clinically used to image enhanced metabolism of cellular components of the plaque including macrophages.8 Other studies have confirmed the relevance of 18F-FDG for plaque imaging in Watanabe rabbits9 and in patients with calcifications of the arterial wall10 or active atherosclerosis.11 These studies are of interest as macrophages play a crucial role in atherogenesis and, particularly, in plaque rupture.12,13
18F-labeled fluorocholine (18F-FCH) has been introduced as a tracer for imaging brain and prostate cancer.14,15 Choline is taken up into cells by specific transport mechanisms, phosphorylated by choline kinase, metabolized to phosphatidylcholine, and eventually incorporated into the cell membrane.16–18 Increased choline uptake has been shown in tumor cells19 and activated macrophages.17 Based on this concept, we have recently demonstrated an enhanced 18F-FCH uptake that correlated with macrophage accumulation as part of an inflammatory reaction after soft tissue infection20 or acute cerebral radiation injury.21
Therefore, we compared the ability of 18F-FCH and 18F-FDG to image murine atherosclerotic plaques ex vivo and validated these results against fat staining.
Materials and Methods
For more details, please see http://atvb.ahajournals.org.
Male atherosclerotic apolipoprotein E knockout (ApoE−/−, C57Bl/6J) mice22 were fed a high-cholesterol diet (1.25% total cholesterol; RD12108; Research Diets) for 2 months starting at the age of 8 weeks. Animals were kept without food for 4 hours before injections of the radiotracers until sacrifice. A subset of wild-type C57Bl/6J mice without atherosclerosis received a normal chow diet. All animal experiments were performed in accordance with our institutional guidelines and approved by the local animal committee.
Ex vivo plaque imaging was performed after injections of 18F-FCH (45.8 to 60.5 MBq; n=5), 18F-FDG (36.7 to 46.4 MBq; n=5), or coinjections of 18F-FCH and 14C-FDG (185kBq; n=3) in 300 μL normal saline into the animals’ tail veins.
Harvesting and Tissue Processing
For determining the uptake of 18F-FCH, 18F-FDG, or 14C-FDG within plaques, ApoE−/− mice were euthanized 20 minutes after injection of the radiotracer(s); additional animals were euthanized 3 hours after 18F-FDG injections. For en face analyses, distal aortae were opened longitudinally. For microscopic examinations, 3 samples of the proximal aorta were embedded in OCT compound (Tissue-Tek, Sakura, the Netherlands) and frozen in isopentane. Serial cross-sections of 10 μm thickness were immediately cut and thaw-mounted on glass slides. Autoradiography and fat stainings were performed on the same samples (n=12 from each animal). For biochemical analyses, aortae from ApoE−/− (n=3) and wild-type mice (n=3) were shock-frozen in liquid nitrogen (LN2) and stored at −80°C.
Choline Kinase Activity and Expression
For more details, please see http://atvb.ahajournals.org.
Plaque areas were assessed via fat staining using Oil-red O. Macrophages were identified on adjacent cross-sections using a rat anti-mouse CD68 monoclonal antibody (Serotec, Clone FA-II, 1:400).
For en face macroautoradiography, aortae were longitudinally opened and exposed on a phosphor imaging screen with 14C standards (for 4 hours with 18F-labeling; for 10d using 14C-labeling). Furthermore, aortic cross-sections of 10 μm thickness were exposed with 14C standards for microautoradiography. The data were scanned (Fuji BAS 1800 II; pixel size, 50 μm) and converted to kBq/cc.
The lesion-to-nonlesion ratio in autoradiographies of longitudinally opened aortae was determined using linear integration of signal intensity (NIH Image J 1.33 software) over corresponding aortic areas taking the mean of 5 measurements from each animal. The corresponding ratio in 3 cross-sections of the proximal aorta was determined after injections of 18F-FCH or 18F-FDG by relating the signal intensity of a region of interest within the plaque to a region without lesion. In addition, the standardized uptake value (SUV) was obtained for each animal by dividing the target tissue radioactivity uptake (kBq/cc) within a region of interest by the injected activity per gram of body weight. Comparisons between en face fat stainings and corresponding autoradiographies were determined by converting the autoradiographies into color-coded images, tracing all positive areas (Analysis Five Docu, SoftImaging System) and correlating the percent positive of the total vessel areas using the Spearman rank correlation test (GraphPad Prism V4). Analogous comparisons were performed on 3 cross-sections of each animal in the proximal aorta after coinjections of 18F-FCH and 14C-FDG (n=3) for autoradiographies, fat, and CD68 stainings.
For more detail, please see http://atvb.ahajournals.org.
An unpaired t test was used to compare results between different groups. Values are given as mean±SD; P<0.05 was considered statistically significant.
18F-FCH Macroscopically Visualizes Murine Atherosclerotic Plaques
En face analysis of atherosclerotic plaques revealed a strong and selective uptake of 18F-FCH (Figure 1A). The mean signal-to-noise ratio of these macroautoradiographies relating radioactivity uptake in plaque-bearing to plaque-free vessel wall was 4.9±2.0 to 1 (n=5). Comparisons of the corresponding en face autoradiographic signals and fat stainings in single plaques demonstrated a highly significant correlation (r=0.842, P<0.0001) and a sensitivity of 84% to detect fat-stained areas by autoradiography. Because 18F-FCH uptake in rodents and men reaches a plateau 20 minutes after radiotracer injection,14,15,20,21,23 we did not study further time intervals for plaque imaging using 18F-FCH.
18F-FDG Visualizes Murine Plaques Less Specifically Than 18F-FCH
In light of previous reports with 18F-FDG to image plaque biology in rabbits9 and patients,8,10,11 we validated the ability of 18F-FDG to image murine atherosclerotic plaques applying the same experimental protocol as described for 18F-FCH; 20 minutes after injection, 18F-FDG uptake on autoradiography correlated poorly with the fat staining of plaques (Figure 1B, top). The mean signal-to-noise ratio of radioactivity uptake in plaque-bearing versus plaque-free vessel wall was 6.0±5.1 to 1 (n=5). The sensitivity of the autoradiography after 18F-FDG injection was only 64% to detect the fat-stained areas. Comparisons of the corresponding en face autoradiographic signals and fat stainings (Figure 1B) documented a lower correlation (r=0.261, P<0.05) than for 18F-FCH.
Given the clinical report describing an interval of 3 hours as the optimal time course for 18F-FDG imaging using PET-CT,8 we investigated whether harvesting aortae 3 hours after intravenous 18F-FDG injections would affect plaque imaging ex vivo. Comparisons between en face autoradiographies and fat stainings 3 hour post radiotracer injections (Figure 1B, bottom; n=3) revealed a slightly better correlation (r=0.476, P<0.001; Table) than after 20 minutes. However, the mean signal-to-noise ratio of radioactivity uptake in plaque-bearing versus plaque-free vessel wall decreased to 2.6±0.9 to 1 (n=3). In addition, because of a high rate of 76% false-negative autoradiographic signals, the sensitivity of the autoradiography 3 hours after 18F-FDG injection was only 57% to detect the fat-stained areas.
18F-FCH Uptake Colocalizes With Plaques Using Microautoradiography
Analyses of cross-sections from the proximal aorta showed that 18F-FCH uptake correlated well with plaques (Figure 2A), whereas 18F-FDG uptake showed a lower sensitivity to identify atherosclerotic plaques (Figure 2B). The maximum of the SUV in the plaque was 1.8±1.0 for 18F-FCH and 2.4±0.5 for 18F-FDG (n=5; not significant [NS]). In addition, the signal-to-noise ratio (plaque-bearing vessel portion to plaque-free vessel wall) was 3.5±0.9 to 1 for 18F-FCH and 2.9±0.5 to 1 for 18F-FDG (NS).
18F-FCH Uptake Colocalizes With Plaque Macrophages
Autoradiographic analyses of aortic cross-sections after intravenous coinjections of 18F-FCH and 14C-FDG into the same ApoE−/− mice showed similar radionuclide uptake in plaque-bearing vessel areas (Figure 3A, top; Table). Under these conditions, the signal-to-noise ratio (lesion to nonlesion) calculated from the SUV was 2.5±0.4 to 1 for 18F-FCH and 2.4±1.1 to 1 for 14C-FDG (n=3; NS).
Staining of adjacent cross-sections for macrophages using an anti-CD68 antibody (Figure 3A, bottom) revealed that 18F-FCH uptake correlated better with fat staining (r=0.740, P<0.0001) and CD68 positive areas (r=0.740, P<0.0001; Figure 3B, top; Table) than 14C-FDG (fat: 0.236, P=0.29 and CD68 staining: 0.352, P=0.11; Figure 3B, bottom), respectively.
18F-FCH Mouse PET
Mouse PET scans obtained 10 to 30 minutes after 18F-FCH injections could not detect a specific tracer uptake in atherosclerotic aortae of ApoE−/− mice but showed an increased unspecific 18F-FCH uptake in metabolically active tissues such as heart and liver. Subsequent en face analyses of these aortae revealed a good colocalization of autoradiographies and fat stainings.
Choline Kinase Expression and Activity Are Unchanged in Atherosclerotic Versus Normal Aortae
As an increased choline uptake in tumor cells has been related to an upregulation of choline kinase and/or choline transport mechanisms,17,19 we compared expression and activity of choline kinase in atherosclerotic versus normal aortae isolated from ApoE−/− or wild-type mice, respectively. Both choline kinase expression and activity were similar. No methods are currently available for investigating choline transport in vivo.
It is now well-recognized that plaque formation is a complex, noncontinuous process and that plaque rupture is not just a matter of size (ie, morphology), but mainly the consequence of plaque vulnerability (ie, biology). Thus, imaging of vulnerable plaques constitutes a great need in cardiovascular medicine.5
In this study, we describe for the first time to our knowledge 18F-FCH as a promising agent for imaging biological properties of murine atherosclerotic plaques using ex vivo macroautoradiography and microautoradiography. At the cellular level, uptake of 18F-FCH correlated significantly with plaque macrophages. These findings extend our previous studies in which we described 18F-FCH for noninvasive imaging of a soft tissue infection in a rat model.20
Activated macrophages have been shown to play a key role in promoting atherogenesis. For example, by infiltrating the intimal layer, they replicate, become foam cells, and express and secrete hazardous inflammatory molecules within the arterial wall. Particularly, their formation of enzymes such as matrix metalloproteinases (MMPs) leads to thinning of the fibrous cap, thereby increasing the vulnerability of plaques.12,13
Enhanced 18F-FDG uptake was recently reported in plaques of atherosclerotic rabbits9 and humans.8,10,11 Furthermore, increased 18F-FDG uptake in unstable plaques colocalized with plaque macrophages,8 suggesting increased metabolic activity. Thus, these authors proposed enhanced 18F-FDG uptake in human plaques as a typical sign of plaque vulnerability. These reports and our previous findings inspired us to perform a side-by-side comparison of en face preparations 20 minutes after 18F-FCH or 18F-FDG injections complemented by cross-sections after simultaneous injections of 18F-FCH and 14C-FDG into ApoE−/− mice. As image acquisition 3 hours after 18F-FDG injection was identified as the optimal time point for visualizing human vulnerable plaques via PET-CT,8 we also investigated ex vivo imaging of murine plaques 3 hours after 18F-FDG administration. Our macroscopic en face measurements demonstrated that uptake of FCH visualizes murine plaques more specifically than FDG 20 minutes or 3 hours after radiotracer injection. In particular, the sensitivity for detecting plaques ex vivo was best for 18F-FCH, lower for 18F-FDG at 20 minutes, and lowest 3 hours after 18F-FDG at injection. In addition, our histological analyses revealed that 18F-FCH uptake correlated significantly with fat and macrophage stainings better than the corresponding correlations of 14C-FDG uptake. These macroautoradiographic and microautoradiographic findings suggest that FCH may even be better than FDG for visualizing murine plaques ex vivo and render 18F-FCH a promising candidate for noninvasive imaging of plaque metabolism.
To be applicable for noninvasive plaque imaging, the radiolabeled particles not only have to accumulate within the target tissue, they also have to be cleared quickly from the circulating blood to provide a sufficient signal-to-noise ratio for scintigraphy or PET imaging. For this purpose, 18F-FCH appears a suitable and promising tool given its rapid blood clearance rate.23 However, our 18F-FCH PET scans of ApoE−/− mice did not allow us to detect a plaque-specific signal. We think that this was caused by: (1) the difficulty to relate the PET signal to a specific anatomic structure; (2) the limited spatial resolution of small animal PET (&1 mm) to detect even smaller murine plaques (&200 to 500 μm); (3) the unspecific uptake of 18F-FCH in metabolically active tissues such as liver and heart; and (4) the moderate SUV of 18F-FCH as suggested by our experiments. However, we speculate that human plaque imaging using 18F-FCH is feasible because of: (1) the opportunity to create hybrid images by combining high spatial resolution morphological imaging using CT or MRI with PET; (2) the considerably bigger size of human compared with murine plaques; (3) the similar SUV for 18F-FDG and 18F-FCH; and (4) the feasibility of human plaque imaging using 18F-FDG.8
Radiolabeled choline is known as a proliferation marker and has been used for imaging brain and prostate cancer.14,15 The increased choline uptake in highly proliferative cells such as tumor cells has been related to an upregulation of choline kinase as well as an increased activity of choline-specific transporters.19,24 We show similar levels of choline kinase expression and activity in normal and atherosclerotic murine aortae. Thus, enhanced 18F-FCH uptake in activated murine plaque macrophages is not caused by changes in choline kinase, but rather by increased choline transport. The rapid uptake (within 30 minutes) of 18F-FCH into prostate cancer14,15 or inflammatory tissues20 supports this notion. Interestingly, choline and FCH are trapped in the cells, whereas the nonmetabolizable choline analogue 18F-deshydroxy-FCH, which also uses the specific transport system, is weakly incorporated in the cells.24 Thus, accumulation of FCH 20 minutes after injection in our study suggests a specific uptake, because nonspecific transport would not lead to intracellular radiotracer accumulation.
Other radionuclide-based approaches for imaging plaque biology have been reported using radiotracers such as technetium (99mTc)-linked, Indium (111In)-linked, or iodine (125I, 131I)-linked compounds.7 Many of the previous studies used macrophage labeling to characterize one important aspect of plaque vulnerability.25 For example, 99mTc-labeled Annexin V visualized apoptotic macrophages in plaques of atherosclerotic rabbits26 and in a small number of patients.27 Similarly, uptake of 125I-linked monocyte chemoattractant protein 1 (MCP-1) revealed an excellent correlation with macrophages in aortic plaques of atherosclerotic rabbits.28 Schäfers et al recently reported on scintigraphic imaging of matrix metalloproteinase (MMP) activity in ligation-induced and cholesterol-induced carotid lesions of ApoE−/− mice using a 123l-labeled MMP inhibitor.29 Similar to our results, the radiotracer signal in these lesions colocalized with activated macrophages. However, the impact of these findings with regard to imaging of murine atherosclerotic plaques has to be interpreted with caution as the type of lesions after complete carotid ligation30 is different from atherosclerotic plaques in terms of pathophysiology, cellular contents, and size.
Overall, given its limited spatial resolution, scintigraphic imaging using single-photon emission CT (SPECT) may severely limit the detection of small plaques. When compared with SPECT, PET is superior in terms of image resolution (4 mm versus 10 mm) and sensitivity. Although even PET-based devices do not have the spatial resolution to provide detailed tissue characterization, the lesions can be detected if there is a sufficiently high target-to-background ratio. Furthermore, the limited spatial resolution can be addressed by hybrid imaging, which combines radionuclide-based visualization of plaque biology with imaging modalities that provide better anatomic detail such as multislice CT2 or MRI.31 For example, PET-CT has been applied successfully at our institution for staging lung cancer32 or assessing myocardial perfusion.33 As mentioned, PET-CT has also been used for imaging plaque metabolism using 18F-FDG.8 Unfortunately, FDG and FCH are taken up into other metabolically active tissues such as myocardium or liver, a problem that currently excludes its use for imaging coronary atherosclerosis. Finally, the advent of catheter-based devices may provide an attractive tool for invasive detection of plaques.34 The proximity of the β-probe close to the target is likely to improve the detection of plaque-associated radioactive signals. Combining this approach with high-resolution morphological imaging such as OCT4 would offer an attractive technique for detailed imaging of plaque biology.
In conclusion, our findings characterize 18F-FCH as a novel agent for imaging relevant aspects of plaque biology. Its colocalization with plaque macrophages may render it an additional marker of vascular inflammation suggesting vulnerability of an atherosclerotic plaque. Its favorable pharmacokinetics with rapid blood clearance as well as the opportunity to visualize 18F-FCH noninvasively by PET-CT may render this approach an attractive tool for risk stratification of atherosclerotic lesions in patients.
This work was funded in part by grants from the European Union G5RD-CT-2001–00532 and by Bundesamt für Bildung und Wissenschaft 02.0057 (C.M.M., T.F.L.), the Hartmann-Müller Foundation (C.M.M., T.v.L.), the Swiss National Science Foundation 3100-068118 (T.F.L., C.M.M.), PP00A-68835 & 3100-68386.02 (P.A.K.), the Swiss Heart Foundation (P.A.K., A.B., C.M.M.), Sassella and Olga Mayenberg Foundation (A.B.), the Spanish MSC grant FIS C03-08 (J.C.L., A.R.d.M.), as well as support by Abbott/Jomed, Germany (C.M.M). We thank Tibor Cservenvàk and Rolf Hesselmann for radiotracer production.
P.A.K. and A.B. contributed equally to this work.
- Received July 5, 2005.
- Accepted November 29, 2005.
Nair A, Kuban BD, Tuzcu EM, Schoenhagen P, Nissen SE, Vince DG. Coronary plaque classification with intravascular ultrasound radiofrequency data analysis. Circulation. 2002; 106: 2200–2206.
Viles-Gonzalez JF, Poon M, Sanz J, Rius T, Nikolaou K, Fayad ZA, Fuster V, Badimon JJ. In vivo 16-slice, multidetector-row computed tomography for the assessment of experimental atherosclerosis: comparison with magnetic resonance imaging and histopathology. Circulation. 2004; 110: 1467–1472.
Takaya N, Yuan C, Chu B, Saam T, Polissar NL, Jarvik GP, Isaac C, McDonough J, Natiello C, Small R, Ferguson MS, Hatsukami TS. Presence of intraplaque hemorrhage stimulates progression of carotid atherosclerotic plaques: a high-resolution magnetic resonance imaging study. Circulation. 2005; 111: 2768–2775.
Tearney GJ, Yabushita H, Houser SL, Aretz HT, Jang IK, Schlendorf KH, Kauffman CR, Shishkov M, Halpern EF, Bouma BE. Quantification of macrophage content in atherosclerotic plaques by optical coherence tomography. Circulation. 2003; 107: 113–119.
Jaffer FA, Weissleder R. Seeing within: molecular imaging of the cardiovascular system. Circ Res. 2004; 94: 433–445.
Rudd JH, Warburton EA, Fryer TD, Jones HA, Clark JC, Antoun N, Johnstrom P, Davenport AP, Kirkpatrick PJ, Arch BN, Pickard JD, Weissberg PL. Imaging atherosclerotic plaque inflammation with [18F]-fluorodeoxyglucose positron emission tomography. Circulation. 2002; 105: 2708–2711.
Ogawa M, Ishino S, Mukai T, Asano D, Teramoto N, Watabe H, Kudomi N, Shiomi M, Magata Y, Iida H, Saji H. (18)F-FDG accumulation in atherosclerotic plaques: immunohistochemical and PET imaging study. J Nucl Med. 2004; 45: 1245–1250.
Ben-Haim S, Kupzov E, Tamir A, Israel O. Evaluation of 18F-FDG uptake and arterial wall calcifications using 18F-FDG PET/CT. J Nucl Med. 2004; 45: 1816–1821.
Boggs KP, Rock CO, Jackowski S. Lysophosphatidylcholine and 1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine inhibit the CDP-choline pathway of phosphatidylcholine synthesis at the CTP:phosphocholine cytidylyltransferase step. J Biol Chem. 1995; 270: 7757–7764.
Spaeth N, Wyss MT, Weber B, Scheidegger S, Lutz A, Verwey J, Radovanovic I, Pahnke J, Wild D, Westera G, Weishaupt D, Hermann DM, Kaser-Hotz B, Aguzzi A, Buck A. Uptake of 18F-fluorocholine, 18F-fluoroethyl-L-tyrosine, and 18F-FDG in acute cerebral radiation injury in the rat: implications for separation of radiation necrosis from tumor recurrence. J Nucl Med. 2004; 45: 1931–1938.
DeGrado TR, Coleman RE, Wang S, Baldwin SW, Orr MD, Robertson CN, Polascik TJ, Price DT. Synthesis and evaluation of 18F-labeled choline as an oncologic tracer for positron emission tomography: initial findings in prostate cancer. Cancer Res. 2001; 61: 110–117.
Schaar JA, Muller JE, Falk E, Virmani R, Fuster V, Serruys PW, Colombo A, Stefanadis C, Ward Casscells S, Moreno PR, Maseri A, van der Steen AF. Terminology for high-risk and vulnerable coronary artery plaques. Report of a meeting on the vulnerable plaque, June 17 and 18, 2003, Santorini, Greece. Eur Heart J. 2004; 25: 1077–1082.
Kolodgie FD, Petrov A, Virmani R, Narula N, Verjans JW, Weber DK, Hartung D, Steinmetz N, Vanderheyden JL, Vannan MA, Gold HK, Reutelingsperger CP, Hofstra L, Narula J. Targeting of apoptotic macrophages and experimental atheroma with radiolabeled annexin V: a technique with potential for noninvasive imaging of vulnerable plaque. Circulation. 2003; 108: 3134–3139.
Ohtsuki K, Hayase M, Akashi K, Kopiwoda S, Strauss HW. Detection of monocyte chemoattractant protein-1 receptor expression in experimental atherosclerotic lesions: an autoradiographic study. Circulation. 2001; 104: 203–208.
Schafers M, Riemann B, Kopka K, Breyholz HJ, Wagner S, Schafers KP, Law MP, Schober O, Levkau B. Scintigraphic imaging of matrix metalloproteinase activity in the arterial wall in vivo. Circulation. 2004; 109: 2554–2559.
Kumar A, Lindner V. Remodeling with neointima formation in the mouse carotid artery after cessation of blood flow. Arterioscler Thromb Vasc Biol. 1997; 17: 2238–2244.
Toussaint JF, LaMuraglia GM, Southern JF, Fuster V, Kantor HL. Magnetic resonance images lipid, fibrous, calcified, hemorrhagic, and thrombotic components of human atherosclerosis in vivo. Circulation. 1996; 94: 932–938.
Koepfli P, Hany TF, Wyss CA, Namdar M, Burger C, Konstantinidis AV, Berthold T, Von Schulthess GK, Kaufmann PA. CT attenuation correction for myocardial perfusion quantification using a PET/CT hybrid scanner. J Nucl Med. 2004; 45: 537–542.