Abstract
RP170, 1-(2-hydroxy-1-[hydroxymethyl]ethoxy)methyl-2-nitroimidazole, is a new hypoxic radiosensitizer. We recently succeeded in labeling this compound with 18F to make 18F-FRP170, 1-(2-fluoro-1-[hydroxymethyl]ethoxy)methyl-2-nitroimidazole. In this study, we attempted to visualize the ischemic but viable myocardium of rats as hot-spot images using 18F-FRP170. We also compared the distribution of 18F-FRP170 with myocardial perfusion, fatty acid metabolism, and glucose metabolism. Methods: In open-chest rats, the left coronary artery was ligated to make an ischemic myocardial model. We evaluated the myocardial accumulation of radiotracers by the double-tracer autoradiography technique. 14C-Iodoantipyrine (IAP), 125I-15-(p-iodophenyl)-3-(R,S)-methylpentadecanoic acid (BMIPP), and 14C-deoxyglucose (DG) were used as markers of myocardial perfusion, fatty acid metabolism, and glucose metabolism, respectively. Approximately 80 small circular regions of interest were placed throughout the left ventricular wall of the midventricular level section. The uptake in each region of interest was expressed as the percentage uptake of the average count in the remote area. We then defined the 18F-FRP170 high-uptake area (H-FRP) as that area where the percentage of FRP was >120% and the 18F-FRP170 low-uptake area (L-FRP) as that area where the percentage of FRP was <80%. Results: On the 18F-FRP170 image of the ischemic myocardium, there was a low-uptake area in the center, which was bounded by a high-uptake area. The percentages of IAP in the H-FRP and the L-FRP were 48.5% ± 5.8% and 2.2% ± 0.2%, respectively; the percentages of BMIPP in the H-FRP and the L-FRP were 46.8% ± 3.1% and 4.3% ± 0.4%, respectively; and the percentages of DG in the H-FRP and the L-FRP were 107.0% ± 6.8% and 4.3% ± 0.4%, respectively. In diabetic rats, the percentages of DG in the H-FRP and the L-FRP were 407.7% ± 14.9% and 72.5% ± 5.0%, respectively, but the remote area on the DG image was insufficiently visualized. Histologic data using 2,3,5-triphenyltetrazolium chloride staining suggest that the high-uptake areas on 18F-FRP170 images reflect ischemic but viable myocardium. Conclusion: We succeeded in visualizing the ischemic but viable myocardium as a hot spot on the 18F-FRP170 image. 18F-FRP170 can be expected to provide important information for determining the necessity of coronary intervention in ischemic heart disease patients, including diabetic patients.
The 2-nitroimidazoles were developed as selective radiosensitizers of hypoxic cells and are used as adjuvants to radiotherapy in the treatment of malignant tumors (1). They have an interesting property—namely, they accumulate selectively in hypoxic cells. The mechanism for the intracellular retention in hypoxic cells is not fully understood. It is believed that 2-nitroimidazoles undergo nitro-reduction with the formation of products that bind to intracellular elements and remain trapped in hypoxic tissues. The enzymatic nitro-reduction required for trapping does not occur unless the cell is viable (2). Therefore, nitroimidazoles increasingly accumulate in the viable area, but not in the necrotic area, of the ischemic myocardium.
In 1981, it was suggested that radiolabeled 2-nitroimidazoles could be used for the direct visualization of tissue hypoxia in tumors (3). 18F-Fluoromisonidazole was the first such radiopharmaceutical developed (4,5), and then many types of radiolabeled 2-nitroimidazoles were synthesized subsequently (6–8). These compounds have been applied to myocardial imaging. Compounds characterized by high lipophilicity were thought to be better for imaging hypoxia because increased lipophilicity results in increased uptake into the myocytes (9). However, the hepatocellular uptake and retention also increases. Moreover, the blood clearance becomes slow and the contrast between the blood pool and the organs of interest decreases (10). These factors complicate in vivo imaging of myocardial hypoxia or ischemia. Recently, efforts have been directed toward reducing lipophilicity (11–15). RP170 (16), 1-(2-hydroxy-1-[hydroxymethyl]ethoxy)methyl-2-nitroimidazole, is a compound that reduces lipophilicity. It was developed and kindly provided by POLA Chemical Industries (Yokohama, Japan). Its octanol–water partition coefficient is 0.094 versus 0.35 for misonidazole and ∼40 for BMS181321 (15,16).
We recently succeeded in labeling RP170 with 18F to make 18F-FRP170, 1-(2-fluoro-1-[hydroxymethyl]ethoxy)methyl-2-nitroimidazole (17). The change in lipophilicity induced by fluorinating is undetermined, but 18F-FRP170 is expected to have low lipophilicity and to show higher-contrast images with a low background level.
In this study, we attempted to determine whether 18F-FRP170 could be used to visualize the ischemic but viable myocardium as a hot-spot image using the double-tracer autoradiographic technique with a perfusion marker. Moreover, we compared the distribution of 18F-FRP170 with those of 125I-15-(p-iodophenyl)-3-(R,S)-methylpentadecanoic acid (BMIPP) and 14C-deoxyglucose (DG). BMIPP is a marker of fatty acid metabolism and is believed to be a more direct index of regional ischemia than the perfusion markers 201Tl and 99mTc-sestamibi. DG is considered to be the gold standard for detecting the ischemic viable myocardium. Finally, we compared the usefulness of this new agent with that of DG under the conditions of diabetes mellitus.
MATERIALS AND METHODS
This study was performed in conformance with the guidelines of the National Institutes of Health for Care and Use of Laboratory Animals and with the approval of the Committee of Animal Experiments at the Cyclotron and Radioisotope Center of Tohoku University. Male Wistar rats (Funabashi Farms, Shizuoka, Japan; body weight, 180–200 g) were used in these studies. They were fed normal rat chow and had tap water ad libitum until they were subjected to the surgical procedure. As an exception, rats in groups 3 and 4 (comparison with glucose metabolism) were subjected to fasting at 12 pm on the day of surgery. We started the surgical procedure at approximately 6 pm. Rats were anesthetized with an intraperitoneal injection of sodium pentobarbital (40 mg/kg) and mechanically ventilated after endotracheal intubation. The right jugular vein was cannulated with a PE-50 polyethylene tube for the injection of radiopharmaceuticals. A left-sided thoracotomy was performed through the fifth intercostal space. A ligature was placed around the left coronary artery (LCA) near its origin using a 6-0 silk suture. To prevent ventricular arrhythmias, 1 mg/kg lidocaine was injected before the LCA was ligated. In this ischemic model, the septum is not subjected to ischemia (18–21). Therefore, we defined the remote area as the nonischemic area in the septum. Three types of double-tracer autoradiography using 18F-FRP170 and other radiotracers were performed as detailed below. Each experimental protocol (protocols 1–4) is summarized in Figure 1.
Experimental protocols. (A) Protocol 1: 18F-FRP170 and 14C-IAP. (B) Protocol 2: 18F-FRP170 and 125I-BMIPP. (C) Protocol 3: 18F-FRP170 and 14C-DG. i.v. = intravenous injection.
In group 1 (n = 3), a comparison with a myocardial perfusion marker was performed using protocol 1. 18F-FRP170 (55.5 MBq), with a specific activity of >26 GBq/μmol, was injected intravenously 30 min after the LCA ligation. Fifteen minutes after the injection of 18F-FRP170, 4-N-methyl-14C-iodoantipyrine ([IAP] 0.185 MBq), with a specific activity of 188 GBq/nmol, dissolved in 1.6 mL 0.9% NaCl, was injected intravenously for 30 s at a constant rate with an infusion pump. Immediately after the end of 14C-IAP infusion, the rats were killed by severing the ascending aorta and the pulmonary trunk. The hearts were excised rapidly and frozen in dry ice.
In group 2 (n = 5), a comparison with aerobic fatty acid metabolism was performed using protocol 2. 125I-BMIPP was obtained from Nihon Medi-Physics Co., Ltd. (Nishinomiya, Japan). Thirty minutes after the LCA ligation, 55.5 MBq 18F-FRP170 and 185 kBq 125I-BMIPP with a specific activity of 710 MBq/mol were injected intravenously. Fifteen minutes after the injection, the rats were killed and the hearts were excised rapidly and frozen in dry ice.
In group 3 (n = 5), a comparison with anaerobic glucose metabolism was performed using protocol 3. 14C-DG was obtained from Amersham International PLC. (Buckinghamshire, U.K.). Thirty minutes after the LCA ligation, 55.5 MBq 18F-FRP170 and 185 kBq 14C-DG with a specific activity of 2.11 GBq/mol were injected intravenously. Fifteen minutes after the injection, the rats were killed and the hearts were excised rapidly and frozen in dry ice.
In group 4 (n = 5), the same protocol as that used for group 3 was performed on streptozotocin-induced diabetic rats. The rats were administered streptozotocin (45 mg/kg of body weight) intravenously 5 d before the surgical procedure. The plasma glucose level was measured by using a Precision QID monitor (MediSense, Bedford, MA), and the plasma insulin level was measured by a rat insulin 125I assay system (Amersham Pharmacia Biotech, Buckinghamshire, U.K.).
Double-Tracer Autoradiography
Approximately 30 frozen heart sections (20 μm thick), taken perpendicular to the long axis of the left ventricle, were prepared. The sections were placed in contact with general-use imaging plates (SR2025; Fuji Photo Film Co., Ltd., Kanazawa, Japan). The first autoradiographic exposure for 1 h detected the distribution of 18F-FRP170. The 18F was allowed to decay for 20 h, and then 14C and 125I images were obtained by a second exposure for 1 and 3 wk, respectively. By single-tracer autoradiography using 14C-IAP, 125I-BMIPP, and 14C-DG, we initially confirmed that these agents were not visualized under conditions applicable for 18F-FRP170 imaging.
One representative autoradiogram per animal with the largest cross-sectional area of the left ventricle was analyzed by a computerized imaging analysis system (Bio-Imaging Analyzer BAS5000; Fuji Photo Film Co.). The image data were recorded as the digitized values (photostimulated luminescence [PSL]) of each pixel (50 × 50 μm) in the analyzing unit of this system (18). Approximately 80 circular regions of interest (ROIs) (the area of each ROI was approximately 0.1 mm2) were placed on the 18F-FRP170 image throughout the left ventricular wall of the midventricular level section. We put the ROIs on the other image at the same sites as the 18F-FRP170 images using the traced film. The uptake values in each ROI were expressed as the autoradiographic intensities ([PS − BG]/A), where BG = PSL of the background and A = area (mm2). We determined the percentage uptake values in each tracer’s image with the average ([PSL − BG]/A) of the remote area assumed to be 100. The percentage of FRP was determined from the 18F-FRP170 image, percentage of IAP from the 14C-IAP image, percentage of BMIPP from the 125I-BMIPP image, and percentage of DG from the 14C-DG image. The mean SD of the percentage of FRP in the remote area of group 1 (n = 3) was 10.19%. Because the value of 2 SD was approximately 20%, we defined the 18F-FRP170 high-uptake area (H-FRP) as that area where the percentage of FRP was >120% and the 18F-FRP170 low-uptake area (L-FRP) as that area where the percentage of FRP was <80%.
Histologic Examination
To confirm that the ischemic but viable myocardium identified by 18F-FRP170 and 14C-DG images is consistent with that determined by a widely accepted histologic method, 2,3,5-triphenyltetrazolium chloride (TTC) (Sigma Chemical Co., St. Louis, MO) staining, we performed an additional experiment using 3 rats. In this protocol, the suture for the LCA ligation was released 30 min after the onset of ischemia. The chest was closed, and rats were allowed to recover for 24 h. We reperfused those hearts because TTC staining requires >2 h after the onset of ischemia to identify precisely necrotic tissue and viable myocardium (22). Those rats were anesthetized again and intubated for artificial ventilation, and the LCA was ligated at the same position as the first ligation. To delineate the risk area, we injected 5 mL 1.0% Evans blue (Sigma Chemical) from the apex of each left ventricle. The hearts were then isolated, washed with saline, and cut into 5 transverse slices. The slices were incubated with 1.5% TTC solution for 15 min at 36°C and then photographed.
Statistical Analysis
Data are presented as mean ± SE. Comparison between the regions was done using ANOVA with a multiple comparison test. P < 0.05 was considered significant.
RESULTS
Figure 2 shows representative autoradiograms of the midventricular heart sections obtained in all protocols. Circumferential profile curves of every 2 radiotracers used in protocols 1, 2, and 3 are shown in Figure 3. Figures 3A, 3B, and 3C correspond to Figures 2A, 2B, and 2C, respectively. Table 1 summarizes the percentages of IAP, BMIPP, and DG in the H-FRP and the L-FRP.
Regional Percentage Uptake Values Compared with Remote Area
Representative double-tracer autoradiograms obtained from each protocol. All coupled autoradiograms were obtained from same slice. In each group, autoradiograms on left show 18F-FRP170 image. Autoradiograms on right show image with 14C-IAP as marker of myocardial perfusion (A), 125I-BMIPP as marker of myocardial fatty acid metabolism (B), and 14C-DG as marker of myocardial glucose metabolism in slice obtained from heart of nondiabetic rat (C) and slice obtained from heart of diabetic rat (D). Autoradiograms on right were expressed at same window level and width.
Circumferential profile curves of 2 radiotracers used in protocols 1, 2, and 3. (A) 18F-FRP170 and 14C-IAP. (B) 18F-FRP170 and 125I-BMIPP. (C) 18F-FRP170 and 14C-DG. ROIs were placed in serial order in midmyocardial layer of left ventricular wall. Sept. = septum; Post. = posterior; Lat. = lateral; Ant. = anterior.
Comparison with Myocardial Perfusion
Figure 2A shows that the distribution of 18F-FRP170 and 14C-IAP in the remote area is homogeneous. On the other hand, the distribution of 18F-FRP170 is increased at the edge of the ischemic area, although the distribution of 14C-IAP is decreased in almost the whole area of the free wall. Representative circumferential profile curves of the percentages of FRP and IAP are shown in Figure 3A. In the path from the septum to the center of the free wall, the increase in 18F-FRP170 uptake starts almost at the same position as where 14C-IAP uptake decreases. The 18F-FRP170 uptake decreases in the center of the free wall below the uptake in the remote area. Table 1 shows that 14C-IAP distribution was significantly lower in the H-FRP and the L-FRP than in the remote area. The percentages of IAP in the H-FRP and the L-FRP were 48.5% ± 5.8% and 2.2% ± 0.2%, respectively (Table 1).
Comparison with Fatty Acid Metabolism
Figure 2B shows that the uptake of 125I-BMIPP decreased in the free wall, especially in the center. According to the representative circumferential profile curves in Figure 3B, the increase in 18F-FRP170 uptake started at almost the same position as where 125I- BMIPP uptake decreased. The 18F-FRP170 uptake decreased in the center of the free wall below the uptake in the remote area. Table 1 shows that 125I-BMIPP has a significantly lower distribution in the H-FRP and the L-FRP than in the remote area. The percentages of BMIPP in the H-FRP and the L-FRP were 46.8% ± 3.1% and 4.3% ± 0.4%, respectively (Table 1).
Comparison with Glucose Metabolism
We did not measure the glucose and insulin levels of rats subjected to autoradiographic study in group 3 because blood sampling may elicit hemodynamic deterioration. However, in our preliminary experiment, we determined those levels in open-chest and artificially ventilated rats (n = 30) in the same feeding state. The concentrations of glucose and insulin were 218 ± 8 mg/mL and 8.29 ± 0.81 ng/mL, respectively.
Figure 2C shows autoradiograms of 18F-FRP170 and 14C- DG. In the left ventricular free wall, there is a low-uptake area that is bounded by the high-uptake area on both autoradiograms. However, the area of increased 18F-FRP170 uptake is slightly broader than that of 14C-DG. According to the representative circumferential profile curves in Figure 3C, the increase in 18F-FRP170 uptake started at almost the same position as where 14C-DG uptake increased. But the decrease in 14C-DG uptake started closer to the remote area than the point where 18F-FRP170 uptake started decreasing. The percentages of DG in the H-FRP and the L-FRP were 107.0% ± 6.8% and 4.3% ± 0.4%, respectively (Table 1)—namely, the percentage of DG in the H-FRP exceeded 100% but did not reach 120%. On all autoradiograms in this protocol, the maximum percentages of FRPs were higher than the maximum percentages of DGs (257.6% ± 16.7% vs. 187.2% ± 22.1%; n = 5, P < 0.01).
Comparison with Glucose Metabolism in Diabetic Rats
The plasma glucose level of the diabetic rats was 368 ± 11 mg/dL and the plasma insulin level was 0.275 ± 0.041 ng/mL. Figure 2D shows the 14C-DG image of diabetic rats, with the autoradiogram expressed at the same window level and width as the 14C-DG image in Figure 2C. The 14C-DG uptake in Figure 2D is very low, especially in the remote area. The distribution of 18F-FRP170 was similar in nondiabetic and diabetic rats. The percentages of DG in the H-FRP and the L-FRP in diabetic rats were 407.7% ± 14.9% and 72.5% ± 5.0%, respectively (Table 1). These values were extremely high compared with any values for nondiabetic rats. The maximum percentage of DG in diabetic rats was higher than the maximum percentage of FRP attributed to the extremely low DG uptake in the remote area because the percentage of DG in the diabetic rats was only a few times as high as the background level. Table 2 shows the autoradiographic intensities expressed by ([PSL − BG]/A]) in all regions of the ischemic hearts of normal and diabetic rats. The 14C-DG uptake values of the ischemic hearts of diabetic rats were extremely low compared with those of nondiabetic rats. However, the 18F-FRP170 uptake values did not show such a remarkable difference between rats with and without diabetes mellitus.
Comparison of Autoradiographic Intensities Between Nondiabetic and Diabetic Rats
Histologic Correlations
Figure 4 shows a representative TTC-stained section of a heart subjected to 30 min of LCA ligation followed by 24 h of reperfusion. Viable myocardium (stained red) is located at the periphery of the risk area, and necrotic myocardium (pale) is at the center. The distribution of viable myocardium identified by TTC staining is similar to that determined by 18F-FRP170 and 14C-DG images.
Representative TTC-stained section of heart subjected to 30 min of LCA ligation followed by 24 h of reperfusion. Dark area represents nonischemic area stained with Evans blue. Viable myocardium in risk area (red) is located at periphery and necrotic myocardium (pale) is at center of risk area.
DISCUSSION
We successfully visualized ischemic but viable myocardium using 18F-FRP170 in rats. It was possible to compare 18F-FRP170 images with the images obtained by the double-tracer autoradiography technique using 14C-IAP, 125I-BMIPP, or 14C-DG. In recent studies of hypoxic markers for ischemic myocardium using the coronary artery occlusion/ reperfusion heart model, the timing of the injection of radiotracers was often discussed. Fukuchi et al. (19) reported that the uptake of BMS181321 in the area at risk was increased significantly only when BMS181321 was injected before ischemia. Johnson et al. (23) reported that focal increased uptake of 99mTc-nitroheterocycle was seen in the risk region in animals injected 5 and 2.2 min before occlusion but not in animals injected 15 min before occlusion and 15 min after reperfusion. On the basis of the known kinetics of delivery and nitroimidazole reduction, these agents have not been considered to be useful for the detection of transient ischemia (2). Therefore, because we used only an acute ischemic heart model without reperfusion, our new hypoxic marker was injected after occlusion. On 18F-FRP170 images of the free wall of the left ventricle, there was an area with higher uptake than that of the remote area and an area with lower uptake than that of the remote area. A comparison of 18F-FRP170 images with 14C-IAP images showed that the 18F-FRP170 H-FRP and L-FRP were located in the low-perfusion area. When cells have an increased accumulation of 18F-FRP170, it means that they are hypoxic but viable, which was confirmed by the TTC staining in this study. Therefore, the H-FRP on 18F-FRP170 images reflects the ischemic viable myocardium. To be exact, we succeeded in visualizing the ischemic but viable myocardium as a hot spot on the 18F-FRP170 image. These areas were located in a relatively mild ischemic area, but the region of the ischemic viable myocardium could not be detected by perfusion images alone.
125I-BMIPP images showed a tendency that was similar to that of previous perfusion images. High- and low-uptake areas on the 18F-FRP170 images were located in the decreased-uptake area on the 125I-BMIPP images. BMIPP is a radioiodinated methyl-branched fatty acid. Under aerobic conditions, the oxidation of fatty acids is the most important source of energy in the form of adenosine triphosphate (24). On the other hand, under ischemic hypoxic conditions, the oxidation of long-chain fatty acids is suppressed, with augmentation of glucose utilization (25–27). Lack of oxygen rapidly depresses β-oxidation, and activated fatty acids are shunted into storage pools as triglycerides and phospholipids (28). In addition, backdiffusion of unmetabolized fatty acids increases significantly because the ischemic tissue is no longer able to activate and trap them. Therefore, the autoradiographic findings of a moderate decrease in 125I-BMIPP uptake in the 18F-FRP170 H-FRP and a severe decrease in the uptake in the 18F-FRP170 L-FRP correspond to cardiac fatty acid metabolism depending on the intracellular partial pressure of oxygen. To our knowledge, no previous study has reported a correlation between the markers of hypoxia and fatty acid metabolism in the ischemic myocardium.
We also studied the correlation between 18F-FRP170 and 14C-DG in the ischemic myocardium using nondiabetic and diabetic rats. In nondiabetic rats, the area with increased 18F-FRP170 uptake was broader and more prominent than the area with increased uptake of 14C-DG. These results suggest that 18F-FRP170 is as useful as DG in detecting ischemic viable myocardium and that the broader hot spots and higher maximum percentage of uptake on 18F-FRP170 images than on 14C-DG images may lead to easier detection of them, especially in the clinical setting using PET.
Much of the increase in glucose metabolism of acute ischemic myocardial cells is caused by the translocation of GLUT4 molecules known as insulin-responsive glucose transporters (29). The conditions or stimuli that are now known to induce glucose transporters in the heart include insulin (30–33), catecholamines (34), increased workload (35), hypoxia (30), and ischemia. Therefore, in this study, the increased uptake of DG in the myocardium is also believed to be caused by the translocation of glucose transporters induced by ischemia or hypoxia (or both). Decreases in the GLUT4 content in the heart have been reported in diabetic rats (36). The glucose transport in response to physiologic stimuli has also been reported to be impaired in diabetic rats (37). In our study, the percentages of DG in the H-FRP and L-FRP were found to be much higher than those in other studies. This finding must be attributed to the remarkably decreased 14C-DG uptake in the remote area. One of the mechanisms for the low DG uptake in the diabetic hearts, especially in the remote area, is thought to be the low level of serum insulin. The uptake of 14C-DG in the remote area of diabetic rats is only several times as high as the background level, which may make it difficult to recognize the anatomic structure of the heart and to determine where the lesion is. On the other hand, the pattern of 18F-FRP170 uptake in the ischemic heart of diabetic rats was the same as that of nondiabetic rats. The 18F-FRP170 images do not seem to be influenced by the presence of diabetes mellitus.
In this study, we showed that direct hot-spot imaging of ischemic but viable myocardium was possible using a newly developed 2-nitroimidazole analog (i.e., 18F-FRP170). This direct hot-spot imaging is potentially of great clinical importance. If the ischemic but viable myocardium can be detected, we could select appropriate candidates for percutaneous interventional treatment or coronary artery bypass grafting. 18F-FDG PET is currently considered to be the gold standard for the assessment of myocardial viability. However, our results suggest that 18F-FRP170 is as useful as 18F-FDG for detecting the ischemic viable myocardium. Furthermore, in this study, the 18F-FRP170 uptake in the ischemic viable myocardium was broader and higher than that of DG. This might enable an easier detection of ischemic viable myocardium than that by FDG in the clinical setting using PET. Finally, 18F-FRP170 may be more useful in diabetic patients than 18F-FDG because the images seem to be independent of the existence of diabetes mellitus. An advantage of this new agent might be that there is no requirement for glucose loading or consideration of the meal plan even in nondiabetic patients. FDG imaging after overnight fasting may possibly show ischemic myocardium as a hot spot. However, we showed previously that stimulated DG uptake in the ischemic border zone still depends on the DG uptake in the nonischemic remote area (21). The FDG uptake in the ischemic myocardium after overnight fasting might be only a relative hot spot but not a prominent hot spot.
This study has some limitations. First, our results showed that the area with increased uptake of 18F-FRP170 was broader (reached to the central ischemic area) and more prominent than the area with increased uptake of 14C-DG. However, we do not know whether this observation indicates that 18F-FRP170 is more sensitive than DG in detecting ischemic viable myocardium or that this new agent overestimates it. We could not perform TTC staining using the same slices of hearts as those used for 18F-FRP170 or 14C-DG imaging because, in this study, reperfusion was mandatory for TTC staining to avoid overestimation of viable myocardium. Further investigation is necessary to address this question, although the difference in the size of ischemic viable areas determined by 18F-FRP170 and 14C-DG was not large. Second, in this study, we used 18F-FRP170 to successfully visualize ischemic but viable myocardium only ex vivo. When we consider the application of 18F-FRP170 to the clinical setting using PET, we have to consider the feasibility of myocardial imaging with 18F-FRP170 in vivo. In this regard, we need to know the ratio of the myocardial count to the blood count, which was not available in this study.
CONCLUSION
In this study, we succeeded in synthesizing a new positron-labeled 2-nitroimidazole analog, 18F-FRP170. This radiopharmaceutical appears to be useful in the same way as DG for detecting ischemic viable myocardium. 18F-FRP170 PET images may be suitable for quantitative evaluation of ischemic viable myocardium and for accurate diagnosis of acute myocardial infarction. However, because experimental studies are not an adequate basis on which to make general conclusions concerning the possible clinical application, further investigations are warranted.
Acknowledgments
The authors thank Drs. Tatsuo Ido, Hiroshi Fukuda, and Kazuo Kubota for their thoughtful suggestions. This work was supported by a grant from the Association for Nuclear Technology in Medicine (Tokyo, Japan).
Footnotes
Received Mar. 27, 2001; revision accepted Sep. 25, 2001.
For correspondence or reprints contact: Shogo Yamada, MD, PhD, Department of Radiology, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-8574, Japan.
E-mail: shogo-y{at}rad.med.tohoku.ac.jp
