Introduction

Atherosclerotic disease is one of the leading causes of morbidity and mortality in western society [1, 2]. Atherosclerotic plaque disruption with superimposed thrombosis is the main cause of the acute coronary syndromes. Acute thrombosis on disrupted atherosclerotic plaque also plays a role in atherosclerosis progression [3]. While atherosclerosis burden remains an important factor in progression of disease, the composition and vulnerability of plaque have emerged as being crucial determinants for the development of the thrombus-mediated acute coronary syndromes [4]. Recent investigations have focused on the central role of macrophages in converting a stable, quiescent plaque to an unstable one with rupture and fissuring [5, 6]. Macrophages have been implicated in weakening the fibrous cap of the plaque due to the secretion of matrix-metalloproteinase (MMP) [7, 8]. MMPs are a family of enzymes that degrade extracellular matrix components, particularly collagen, elastin, and proteoglycans, thereby possibly serving to thin the collagen skeleton of the fibrous cap, leading to plaque rupture and fissuring [9].

Imaging techniques that can characterize the plaque and its components will potentially allow risk stratification in asymptomatic, as well as symptomatic, patients with coronary artery disease and potentially select the most appropriate therapies to help attenuate this risk. In particular, quantification of macrophage burden might help to assess plaque vulnerability and risk for rupture. A positron emitting radiotracer, fluorine-18-2-deoxy-2-glucose ([18F]FDG), an analog of d-Glucose, is a useful marker of high metabolic activity commonly found in tumor-associated macrophages [10, 11]. We and others have shown the feasibility of FDG PET imaging to assess macrophage content [12–14]. Recently, a small study has demonstrated attenuation of plaque inflammation in humans with simvastatin utilizing this imaging technique [15]. The purpose of this study was to assess progression and/or regression of macrophage content in atherosclerotic plaques in an experimental rabbit model using this non-invasive technique.

Methods

Experimental model

A total of eight New Zealand white rabbits were studied after atherosclerosis induction. Figure 1 illustrates the timeline of the experiment. Induction of atherosclerosis was achieved with combination of a moderate atherogenic diet (0.2% cholesterol) for 9 months and double balloon denudation of the aorta 1 week and 13 weeks after diet initiation. At the end of the induction period, the rabbits were randomized to one of two treatment groups: the progression group that received the same atherogenic diet for six more months (n = 4), and the regression group (n = 4) that received a normal diet for 6 months. FDG PET imaging was performed at the commencement and end of the randomized period, after which all animals were sacrificed. The pathological correlation between FDG-PET and macrophage content in these groups has been published previously [14].

Fig. 1
figure 1

Timeline of experiment. All eight rabbits underwent atherosclerosis induction with 9 months of atherogenic diet (0.2% cholesterol) and balloon denudation of the thoracic aorta at week 1 (W1 BD) and week 13 (W13 BD). The rabbits were randomized to a further 6 months of either the same atherogenic diet or normal diet. FDG PET imaging was performed at 9 months and 15 months on all rabbits

Full size image

Balloon denudation

All procedures were performed under general anaesthesia by intramuscular injection of ketamine (20 mg/kg, Fort Dodge Animal Health, Fort Dodge, IA) and xylazine (10 mg/kg, Bayer Corporation, Shawnee Mission, KA). Aortic denudation of the aorta from the aortic arch to the iliac bifurcation was performed by withdrawal, with moderate resistance, of a 4Fr Fogarty embolectomy catheter introduced through the iliac artery and passed into the aortic arch. Catheter insertion and inflation were repeated for four passes, after which the catheter was removed, the femoral artery tied, and the incision closed as previously described [14, 16–18]. All experiments were approved by the Mount Sinai School of Medicine animal management program, under accreditation from the American Association for the Accreditation of Laboratory Animal Care (AALAC).

FDG-PET imaging of rabbits

Immediately prior to the FDG-PET imaging, all rabbits underwent magnetic resonance imaging (MRI) for the purpose of localization of the anatomical structures on the PET images. After fasting the previous night, the rabbits were anesthetized as described above. Then, they were immobilized with a body-fitting thermosetting plastic holder compatible with both the MRI and PET scanning systems, thus assuring the exact positioning of the rabbits for both imaging studies. The rabbits were laser-aligned on the xiphisternum in both scanning modalities. After initial gradient echo scout images to identify the thoracic aorta, 2D-time of flight imaging was performed to exactly localize the aorta. Thereafter, the anesthetized rabbits were immediately taken to the PET scanner in the holders. Imaging was then performed using a GE 2048-plus brain dedicated PET scanner. The axial field of view was 10.5 cm and the measured in-plane resolution was 4.5 mm in the center of the transverse planes. A 30-min transmission scan was performed for the whole descending aorta of the rabbit. Each rabbit was injected via a marginal ear vein with 1–2 mCi of FDG. FDG (T 1/2 = 110 min) was purchased form PET-NET System (New York, NY). The PET imaging was performed 30 min after injection. A 30-min emission scan was then obtained. The data was reconstructed with standard filtered back projection and attenuation correction. Using an in-house software system, each PET slice was co-registered with its corresponding MRI slice guided by anatomical landmarks and the aortic arch. Thus anatomical structures identified by MRI were correlated with the FDG uptake by a process of image superimposition, as previously described (Fig. 2); [14].

Fig. 2
figure 2

Demonstration of the co-registration of PET and MR images allowing the exact determination of FDG uptake in the aorta. These images are from the same animal showing MRI (a), the resultant fusion (b) and FDG PET (c). The yellow arrows highlight the position of the aorta and the white arrows the position of the heart

Full size image

Blood sampling

Arterial blood samples were collected from an auricular artery at 15, 30, 45, and 60 min post-injection of FDG. Blood samples were weighed. The radioactivity was counted in a gamma counter (Minaxi Auto-Gamma 5000 Series Gamma Counter, Packard, USA) and decay corrected, as previously described [14].

PET image and data analysis

The PET images were transferred to an IBM-compatible computer for further analysis. Aortic uptake was expressed as concentration (μCi/ml), and for the correction of blood pool activity, FDG uptake was expressed as the ratio of aortic uptake-to-blood radioactivity, as previously validated [14]. Slices were 6 mm apart by taking a 5 × 5 mm2 ROI. Serial comparisons were made with FDG uptake using the techniques described. A total of 25 consecutive axial slices, beginning from the distal aortic arch, were analyzed in the descending aorta pre and post dietary modification per animal. This resulted in a total of 100 pairs of data for each group of rabbits.

All FDG PET analyses were performed by a single operator blinded to the dietary randomization.

Statistical analysis

Data is presented as mean ± standard error of the mean. To account for multiple samplings at different sites from the same rabbit, the “svy: regress” procedure was used (STATA 9.1, College Station TX). Paired students t-testing was performed on the matched aortic slices pre and post dietary modification (Statview, Abacus Corp) and the data graphically presented in “box and whisker” formats. A P value < 0.05 was considered statistically significant.

Results

Serum lipid profiles

The serum total cholesterol at baseline was normal (34 ± 23 mg/dl) but was significantly increased (949 ± 141 mg/dl) at the end of atherosclerosis induction at 9 months. The serum total cholesterol at 15 months in the progression group and the regression group were 936 ± 340 and 27 ± 6 mg/dl, respectively.

Histopathology

The combination of the atherogenic diet for 9 months and balloon denudation of the aorta induced a significant thickening of the arterial wall due to the increase in lipid and fibrotic components. The induced atherosclerotic lesions were rich in macrophage areas as indicated by RAM-11 staining (Fig. 3).

Fig. 3
figure 3

In vivo axial PET image showing the uptake of FDG in the thoracic aorta (black arrow) in a control rabbit (a), in a rabbit with mild atherosclerosis (b) and in a rabbit with more advanced atherosclerosis showing a higher uptake of FDG (c). Corresponding histopathology stained with RAM-11 of a, b, c showing no macrophages (d) some macrophages (e) and abundant macrophages (f), respectively

Full size image

[18F]FDG uptake

Using all of the 25 aortic slices pre and post dietary modification per animal, there were significant differences in aortic uptake-to-blood radioactivity value between the progression and regression groups at commencement (0.57 v 0.67, P = 0.003) and end of randomization (0.68 v 0.53, P < 0.001) as determined by unpaired t-testing.

There was a significant increase in the aortic uptake-to-blood radioactivity value in the progression group after a further 6 months of cholesterol feeding (0.57 ± 0.02 to 0.68 ± 0.02, P = 0.001) (Fig. 4).

Fig. 4
figure 4

“Box and Whisker” plot comparing the results of the aortic-to-cardiac uptake activity ratios, pre- and post-continued cholesterol feeding in the atherosclerosis progression group. We can appreciate the significant increase in values representing increased FDG PET uptake and thus increased macrophage content from the pre to post continued cholesterol feeding. The error bars (“whiskers”) represent the 10th and 90th percentiles, respectively, and the top and bottom of the box the 25th and 75th percentiles, respectively. The line in the box represents the median value of the aortic-to-cardiac uptake activity ratios for the given group

Full size image

Similarly, there was a corresponding fall in the aortic uptake-to-blood radioactivity values in the regression group after returning to a normal chow diet for 6 months. (0.67 ± 0.02 to 0.53 ± 0.02, P < 0.0001) (Fig 5).

Fig. 5
figure 5

“Box and Whisker” plot comparing the results of the aortic-to-cardiac uptake activity ratios, pre- and post- cholesterol feeding in the atherosclerosis regression group. We can appreciate the significant decrease in values representing FDG PET uptake and thus decreased macrophage content from the pre to post continued cholesterol feeding. The error bars (“whiskers”) represent the 10th and 90th percentiles, respectively and the top and bottom of the box the 25th and 75th percentiles respectively. The line in the box represents the median value of the aortic-to-cardiac uptake activity ratios for the given group

Full size image

Discussion

We have demonstrated, the ability of [18F]FDG PET to monitor regression and/or progression of macrophage content in an experimental model.

Radionuclide imaging of atherosclerosis uses the pathophysiology of the atherosclerotic process to deliver radiopharmaceuticals to the lesion [19]. Therefore, different constituents of atherosclerosis, including radiolabeled lipoproteins [20, 21], monocytes [22], smooth muscle cells [23], platelets [24, 25] and fibrinogen [26] have been utilized as potential plaque-imaging agents. However, in vivo detection has been difficult to achieve for different reasons. Among these reasons, are the facts that the absolute and specific uptake of radiotracer in atherosclerotic lesions is low and that the blood-pool activity is very high due to slow plasma clearance of the radiotracer [19].

The most important mechanism responsible for the sudden and unpredictable onset of acute thrombosis is plaque rupture [2, 27]. The risk of plaque rupture depends on its composition rather than its size. Ruptures occur preferentially in plaques containing a soft, lipid-rich core that is covered by a thin and inflamed cap of fibrous tissue [4]. Compared with intact caps, the ruptured ones usually are thinner and contain less collagen, fewer smooth muscle cells and many more macrophages. Disrupted fibrous caps usually are heavily degraded and infiltrated by macrophages foam cells [5, 9]. These macrophages are activated, indicating ongoing inflammation at the site of plaque disruption [28]. In human coronaries, there is a correlation between plaque fibrous cap macrophage density and the presence of thin cap fibrous atheroma as detected by optical coherence tomography [29]. There is also a correlation between plaque macrophage density and symptom severity in patients with acute coronary syndromes [30].

Visualization of the vessel wall and the plaque itself rather than the lumen is necessary for the identification of such vulnerable atherosclerotic plaques. Different imaging modalities for assessing plaque composition including magnetic resonance (MR) imaging, angioscopy, vascular ultrasonography and infrared imaging analysis have been developed to discriminate wall components in normal and atheromatous arteries [31–34]. Magnetic resonance techniques appear to be able to non-invasively identify and discriminate the components of complex atherosclerotic lesions both in ex vivo and in vivo settings [33, 34]. However, no non-invasive imaging technique has been shown to identify in vivo macrophages until recently [12–14]. Previous attempts to detect activated macrophages utilized 111In–IgG. Although this was a valid tracer for detection of activated macrophages in tissues ex vivo, it was impossible to detect lesions in the aorta in vivo in rabbits because of a relatively low target-to-background ratio [35].

Fluorine-18-2-deoxy-2-glucose ([18F]FDG) has been shown to be a useful tumor-detecting agent [11, 36]. The mechanism of accumulation of this tracer into the malignant tissue is due to the enhanced rate of glucose utilization by neoplastic cells [37]. Due to increased metabolic demand for glucose, the hexokinase (a key enzyme for glycolysis) activity is increased. 2-Deoxy-2-glucose has been shown to be a substrate for hexokinase [38]. Therefore 2-deoxyglucose analog [18F]FDG is a particularly appropriate imaging agent for tumors. Recent work using microautoradiography has demonstrated that macrophages showed a high uptake of [18F]FDG [10].

In this study, uptake of FDG was assessed in the same rabbit at two time points and a serial comparison between the same atherosclerotic aortic segments made for each rabbit. We were able to monitor and identify significant changes in FDG uptake between the progression and regression groups. Although this may reflect serial changes in macrophage activity alone, we and others have previously shown FDG uptake to correlate with macrophage content [12, 14].

Limitations

Due to the lack of availability of a PET/CT system, MR imaging was used to co-register FDG uptake with the corresponding anatomical sites. We have validated this technique previously [14], but potentially PET/CT systems would allow a simpler and potentially more accurate co-registration to be performed. This study included a relatively small number of animals. Although we have identified significant changes in FDG uptake between the groups, statistical regression to the mean cannot entirely be excluded. A cross-over design may have added further to the potential of FDG PET in serial monitoring of changes in this atherosclerotic model.

We were not able to quantify plaque burden in this study as the MRI images were used purely for localization. However, we have previously demonstrated atherosclerotic burden progression and regression utilizing high resolution MRI with histopathological correlation in the same animal model [17].

Summary

[18F]FDG-PET scanning in this animal model of atherosclerosis has allowed us to serially monitor macrophage content. Thus FDG PET could potentially be used to serially monitor changes in atherosclerotic plaque macrophage content in response to therapies such as lipid lowering over time.