Myocardial Kinetics of a Putative Hypoxic Tissue Marker, 99mTc-Labeled Nitroimidazole (BMS-181321), After Regional Ischemia and Reperfusion
Background A new nitroimidazole complex, 99mTc-propylene amine oxime-1,2-nitroimidazole (BMS-181321), has been developed to allow the positive imaging of hypoxic myocardium by standard gamma camera techniques.
Methods and Results To determine the myocardial kinetics of BMS-181321 during myocardial ischemia and reperfusion, seven open-chest swine were prepared according to a model of extracorporeal coronary perfusion in which left ventricular wall thickening (percent end-diastolic thickness) and substrate use in the left anterior descending (LAD) region ([14C]palmitate and [3H]glucose infusions) were determined. Measurements were obtained at baseline, during 40 minutes of ischemia produced by reducing flow in the LAD distribution by 60%, and during 70 minutes of reperfusion. Three aerobic control hearts were also studied in which LAD blood flow was not reduced. Regional coronary circulation was further assessed in all hearts by use of radiolabeled microspheres injected during ischemia. BMS-181321 (20 to 30 mCi) was injected after 30 minutes of ischemia, and its myocardial uptake was assessed by dynamic planar gamma imaging. Ischemia was associated with declines in fatty acid metabolism (15±11 μmol · h−1 · g dry wt−1, mean±SEM), systolic wall thickening (20±6%), and myocardial oxygen consumption (3±1 mL · min−1 · 100 g−1) and an increase in exogenous glucose utilization (75±13 μmol · h−1 · g dry wt−1). Systolic wall thickening recovered by only 8±3% with reperfusion. Initial distribution of BMS-181321 in the aerobic hearts appeared homogeneous. Washout from the ischemic and reperfused LAD bed was slower than the aerobically perfused LAD bed in the control group (t1/2=136±1 versus 80±1 minutes, P<.05), allowing visualization of the LAD region during reperfusion. Tissue activity of BMS-181321 was inversely related to LAD blood flow during ischemia (r=−.68±.05), and the ratio of BMS-181321 in the LAD region versus normal myocardium was 1.7±0.2. Control swine lacked regional deposition of the tracer in the normally perfused LAD distribution.
Conclusions Thus, acute regional ischemia in these studies was visualized as an increase in retention of BMS-181321, suggesting its applicability in the imaging of clinical conditions of myocardial hypoperfusion.
The 2-nitroimidazole misonidazole has been used as a sensitizer for radiation therapy for tumor cells in which misonidazole localizes in hypoxic cells. The uptake of misonidazole in regions of tissue hypoxia prompted the development of [18F]fluoromisonidazole to image hypoxic tissue. Imaging of ischemic tissue with [18F]fluoromisonidazole has also been performed with uptake of tracer in ischemic regions of the heart and brain.1 The clinical application of imaging with [18F]fluoromisonidazole, however, is constrained by its dependence on cyclotron production of [18F]fluorine as well as imaging with positron emission tomography. To lessen these constraints, a 2-nitroimidazole compound, BMS-181321, has been synthesized that complexes 99mTc.2 Isolated rat heart studies have demonstrated a twofold greater retention of BMS-181321 with hypoxia than with normoxia.3 Collateral studies in brain using BMS-181321 have also shown selective retention in ischemic regions as derived from reports using autoradiography in rat brain and tomographic imaging in cat brain after middle cerebral artery occlusion.4 To study the kinetics of BMS-181321 in ischemic myocardium, intact, working, extracorporeally perfused open-chest pigs were subjected to anterior wall ischemia and reperfusion. Changes in myocardial metabolism, mechanical function, and perfusion were measured and related to simultaneously acquired dynamic imaging of myocardial BMS-181321 radioactivity, ex vivo imaging of short-axis slices of the heart, and radionuclide activity in tissue sections.
The experimental protocol has been extensively described elsewhere5 6 7 8 9 and was approved by the University of Wisconsin–Madison Research Animal Resource Center Committee. Swine were premedicated with ketamine 11 mg/kg, acepromazine 0.8 mg, and atropine 1.1 mg/kg IM. Anesthesia was induced with sodium pentobarbital 25 mg/kg IV. The pigs were ventilated via a tracheostomy with 100% oxygen, and the anesthetic state was maintained throughout the perfusion trials with α-chloralose (1.5-g bolus, 0.5 g qh IV) and morphine sulfate (45 mg qh SC). The anterior rib cage was removed, and a pericardial sack was made. A dual-tip Millar pressure catheter was inserted into the left ventricle from the right carotid artery to monitor aortic and left ventricular pressure. Femoral artery and vein were cannulated to serve as the arterial source for the coronary extracorporeal perfusion circuit and as an access site for the return of coronary venous drainage, respectively. The three coronary arteries were cannulated separately. Pressure in the coronary arteries was matched to the systemic pressure by adjustment of extra corporeal perfusion. The great cardiac and left anterial descending (LAD) veins were cannulated for venous blood sampling, and a side port was attached to the LAD arterial line for LAD arterial blood sampling. Thickness crystals were placed in the LAD and circumflex perfusion beds. Seventeen pigs (90 to 110 lb) were instrumented.
BMS-181321 was obtained in kit form from Bristol-Myers Squibb Pharmaceutical Research Institute. [99mTc]TcO4− (50 mCi) in saline was added to the ligand vial at room temperature and mixed. After the compound was dissolved, the reductant stannous DTPA (Techneplex, Bristol-Myers Squibb Pharmaceuticals) was added. The reaction mixture was allowed to stand for 10 minutes before injection and was used within 30 minutes of synthesis. Radiochemical purity was confirmed by thin-layer chromatography performed within 1 hour after injection of the compound. For thin-layer chromatography, ethanol was spotted onto a strip of Gelman solvent saturation pad, followed immediately by spotting with BMS-181321. The strip was developed with diethyl ether, and the origin and solvent-front activities were counted to determine radiochemical purity. A minimum purity level of 90% was accepted prospectively. Of the swine studied, one pig was excluded because of a purity level <90%. In the remaining swine, the average purity level was 92.0±0.7% (range, 90% to 95%).
Pigs were divided into an ischemic/reperfused group and a control group. After a baseline period of 30 minutes, the ischemic/reperfused group was subjected to a 40-minute ischemic period with a 60% flow reduction in LAD flow, which was then followed by a reperfusion interval at baseline aerobic flows for 70 minutes. For the control group, LAD flow was maintained at baseline aerobic levels throughout the entire 140-minute experimental period. Hemodynamic and mechanical parameters of heart rate, systemic pressure, coronary perfusion pressure, and wall thickening were acquired at 10-minute intervals. LAD arterial and venous oxygen saturations were obtained for calculating myocardial oxygen consumption. Substrate utilizations were determined for exogenous glycolytic flux with steady-state infusions of [3H]glucose (DuPont NEN Products, Inc; 120 μCi; specific activity, 16 000 dpm/μmol) and for fatty acid oxidation with infusions of [14C]palmitate (DuPont NEN Products, Inc; 50 μCi; specific activity, 111 000 dpm/μmol). Both exogenous labelings were continued for 110 minutes into the LAD circulation, ending 30 minutes before completion of the experiment. Dilution of LAD venous effluent was determined by indocyanine green infusion. LAD arterial and venous samples for calculation of substrate utilization were obtained during baseline, ischemia, and reperfusion (at 20, 30, 40, 50, 60, 70, 80, 90, and 110 minutes).
At t=50 minutes, 20 minutes into ischemia, 20 to 30 μCi of 141Ce-, 113Sn-, or 103Ru-labeled microspheres (15 μm, DuPont NEN Products, Inc) was injected into the femoral arterial line just before the mixing chamber for the three coronary arterial perfusion lines. An arterial reference sample was withdrawn from a sampling port after the mixing chamber at a rate of 4.94 mL/min. At t=60 minutes, 30 minutes into ischemia, [99mTc]BMS-181321 (20 to 30 mCi) was injected into the mixing chamber. Dynamic imaging of the heart was made with a planar gamma camera (General Electric MaxiCam) and a pinhole collimator. The camera was oriented in a 30° right anterior oblique projection. A universal gamma camera interface, developed in-house,10 used standard Nuclear Instrumentation Module–based electronics and Computer-assisted Measurement and Control hardware to acquire digital image data. A 256×256 matrix was used for dynamic image acquisition for all of the experiments. Image acquisition and display were done on a MacIntosh Quadra 950 computer (Apple Computer) with Digital Image Processing Station software (Hayden Imaging Processing Group). The dynamic frame rates were 30-second frames for 10 frames and 120-second frames for 35 frames.
At the end of the protocol, pigs were euthanatized with a lethal dose of pentobarbital. The LAD perfusion bed was stained with india ink injected through the LAD cannula. Since the proximal LAD was tied off during the insertion of the cannula, no reflux of india ink occurred down the circumflex artery. Hearts were then sectioned into 7 to 9 short-axis slices and imaged on the face of a parallel-hole collimator of the gamma camera for 20 minutes.
After the slices were photographed in color, they were cut into epicardial and endocardial sections of 1 to 1.5 g and counted in a gamma counter (Cobra, Packard Instrument Co) for 99mTc and microsphere counts.
Three pigs were studied in the control group. Of the 14 pigs instrumented in the ischemic/reperfused group, 7 pigs survived throughout the perfusion protocol and were available for complete analysis, while 6 others died during the ischemic period, and 1 additional pig was excluded because of impurity of BMS-181321.
Fatty acid oxidation was measured in terms of [14C]CO2 production according to the equation
where Δ[14C]CO2 is the arterial−venous difference in [14C]CO2, QLAD is the LAD coronary artery flow, K is the dilution factor, ASA is the arterial specific activity of [14C]palmitate, and LAD dry wt is the dry weight of the LAD perfusion bed.7 8
Exogenous glucose utilization was calculated as the rate of [3H]H2O production according to the equation
where Δ[3H]H2O is the arterial−venous difference in [3H]H2O, Hct is hematocrit, and ASA is the specific activity of [3H]glucose.6
BMS-181321 washout analysis was performed on the dynamic images, with regions of interest drawn on a late frame over the anterior wall in the ischemic/reperfused group and control groups. The regions of interest were applied throughout the dynamic study to determine washout activity.
The fractional retention of BMS-181321 was calculated from the activity of BMS-181321 in the tissue samples. Efficiency of the gamma counter was determined from counting 99mTc aliquots of known activity. Dilutions of the aliquots were performed to adjust the activity to the range of the tissue samples.
Data are presented as mean±SEM for each variable. ANOVA with Dunnett’s test11 was used to compare changes in systolic thickening, myocardial oxygen consumption, and substrate utilization during ischemia and reperfusion to control levels. Washout rates, myocardial blood flow, epicardial/endocardial return of blood flow, and BMS-181321 were compared by two-tailed Student’s t test for paired or unpaired data, depending on the comparison. A significance level was defined for values of P<.05.
In the seven swine undergoing reduction in LAD perfusion (ischemic/reperfused group), the decline in anterior wall perfusion was accompanied by a reduction in systolic thickening, consistent with an ischemic response (Fig 1⇓). After restoration of flow, systolic thickening improved but did not return to baseline, suggesting of mechanical stunning of the anterior wall. As expected, no significant changes in systolic function occurred in the circumflex region with the maintenance of aerobic perfusion of the myocardium (Fig 1⇓). The absolute systolic thickening was less in the circumflex bed throughout the experiment, probably due to the more basal placement of the crystals. Metabolic changes consistent with an ischemic response were also present during the 60% decrease in LAD flow. A significant decline in myocardial oxygen consumption was noted during hypoperfusion that turned toward baseline during reperfusion (Fig 2⇓). Similarly, an increase in glucose utilization occurred during ischemia, from aerobic values of 2±1 to 76±12 μmol · h−1 · g dry wt−1 (P=.001, Fig 3⇓). A trend toward a reduction in fatty acid utilization, from 24±12 to 10±3 μmol · h−1 · g dry wt−1 (P=.09), was also observed. During reperfusion, there was metabolic recovery with both a decline in glucose utilization and an increase in palmitate utilization. These results are similar to previous studies that used this experimental preparation and protocol.5 6 7 8 9
Three swine were studied as an aerobic control group to determine any changes in oxygen consumption and substrate utilization at a constant LAD flow and to compare washing rates of BMS-181321 in the anterior wall. In this group, there were no significant changes in LAD mechanical function (Fig 4⇓), oxygen consumption (Fig 2⇑), or the utilization of palmitate and glucose (Fig 3⇑).
Dynamic imaging of the myocardium at aerobic flows revealed initial diffuse uptake of BMS-181321 throughout the hearts, followed by washout of tracer. Hepatic uptake of the tracer was progressive during the imaging period. Region-of-interest analysis was performed on the anterior wall for the ischemic pigs and the control pigs. A biexponential decline in activity was seen in all three regions, with rapid washout initially, followed by a slower phase (Fig 5⇓). The data were fitted to a biexponential function to calculate the half-time for the fast component (tr) and the slow component (ts):
No difference in tr was present between the ischemic and control anterior walls (2.6±0.3 versus 1.9±0.2 minutes). A significant prolongation in the second component, ts, was present in the anterior wall of the ischemic/reperfused group compared with the anterior wall in the control group (136±1 versus 80±1 minutes, respectively, P<.05). To determine the relative size of the rapid and slow washout components, the ratio of the intercepts of each component (m1/m2) was calculated. The initial fast component was much larger than the slow component in the anterior wall of the control group: ratio, 31±6. This ratio decreased significantly with ischemia to 12±2, P<.05 compared with the anterior wall in the control group. Region-of-interest size for the analysis of the dynamic study was similar for the two groups, 529±74 pixels for the ischemic/reperfused group and 481±220 pixels for the control group. At the end of the imaging protocol, the LAD bed was readily discernible from the rest of the myocardium in the ischemic/reperfused group (Fig 6⇓). An excellent ratio of heart to lung activity was present in the last dynamic frame, with a ratio of 3.1±0.3 for the ischemic/reperfused group and 2.4±0.4 for the control group. Liver uptake was present, with an average ratio of heart to liver activity of 0.58±0.10 for the ischemic/reperfused group and 0.42±0.12 for the control group in the last dynamic frame.
Imaging of short-axis slices confirmed the regional localization of BMS-181321 (Fig 7⇓). The dynamic image analysis and washout results agreed with the tissue sections, in which the ratio of BMS-181321 in the risk region to the normal region (ie, the ratio of average counts per gram in the risk and normal regions) was 1.7±0.2. Microsphere blood flow data confirmed the decrease in myocardial perfusion in the risk region during ischemia (0.62±0.04 versus 1.27±0.10 mL · min−1 · g−1, risk region versus normal region, P<.01). Hypoperfusion of the risk region was also associated with a greater decrement in endocardial blood flow, increasing the ratio of epicardial to endocardial myocardial blood flow (1.78±0.10 versus 1.16±0.05 for the normal bed, P<.01). The aerobically perfused control LAD bed had no difference in mean myocardial blood flow in the risk region (1.25±0.14 mL · min−1 · g−1) compared with either the normal bed in the control group or the normal bed of the ischemic/reperfused group. The ratio of epicardial to endocardial blood flow in the aerobically perfused risk region (1.08±0.01) was also similar to the normal regions of the control group and the ischemic/reperfused group. The relation of BMS-181321 activity to myocardial blood flow in the tissue sections is shown in Fig 8⇓ for the seven swine from the ischemic/reperfused group. In these animals, BMS-181321 tissue activity (normalized counts per gram) correlated inversely with myocardial blood flow during ischemia in the LAD region, with a mean correlation coefficient for the seven animals of −.68±.05, P<.05. The correlation coefficient for the pooled data for the seven animals in the LAD region was −.48, P<.001. There was no correlation of BMS-181321 activity with blood flow in the normal bed, which remained aerobically perfused during ischemia, with a mean correlation coefficient of .02±.13. (The correlation coefficient for the pooled data was .10.) Similarly, no correlation was seen in the control group of BMS-181321 activity with blood flow in either the risk region, −.18±.11, or the normal region, −.07±.20. Linear regression analysis of the seven swine from the ischemic/reperfused group was performed. For the individual regression analyses, the mean slope was −.39±.05 normalized activity · mL−1 · min−1 · g−1 for the risk region versus .00±.04 for the normal region (P<.001). Similarly, the y intercept was .88±.05 normalized activity for the risk region versus .34±.05 for the normal region (P<.001). The linear regression for the pooled data showed similar relations, with a slope of −.33 normalized activity · mL−1 · min−1 · g−1 and an intercept of .81 normalized activity for the risk region and .04 and .29 for the normal region, respectively. Fig 8⇓ highlights the epicardial/endocardial gradient of BMS-181321 activity in the ischemic region, with greater activity in the endocardium than the epicardium. For the ischemic/reperfused group, the epicardial/endocardial ratio of BMS-181321 declined significantly in the risk region compared with the normal region (0.83±0.02 versus 1.05±0.02, P<.01). Such a difference was not present in the control group, in which the ratio of activity in the risk region (1.05±0.02) was similar to that in the normal region (1.10±0.03).
The tissue counts of BMS-181321 were corrected for the efficiency of the gamma counter and summed for the risk region, the adjacent aerobically perfused region, and all of the right and left ventricles for the ischemic/reperfused group. The retention of BMS-181321 in the right and left ventricles was 0.93±0.08% of the injected dose. The maximal retention of activity in the risk region on a per-gram basis, 0.015±0.002%, was significantly greater than the maximal retention of the normal region, 0.007±0.001%, P<.01. The mean retention in the risk region was 0.008±0.001% of the injected dose per gram of tissue versus 0.005±0.001% in the normally perfused region (P<.05).
For this study, we used an open-chest extracorporeally perfused swine preparation to study the myocardial kinetics of the nitroimidazole BMS-181321 in the setting of myocardial ischemia. The extracorporeal perfusion allowed for excellent control of regional myocardial perfusion and selective administration of labeled glucose and palmitate for determination of regional metabolic changes. The preparation has been used extensively for the study of changes in metabolism with ischemia and reperfusion.5 6 7 8 9 All three coronary arteries were cannulated to permit the infusion of BMS-181321 into all three perfusion beds. Since the BMS-181321 was injected into the mixing chamber before the perfusion pumps, the amount of BMS-181321 presented to the perfusion bed for each coronary artery was proportional to the coronary flow, similar to an intravenous injection of the tracer. The intracoronary injection of tracer, however, means that all of the injected dose is delivered first to the myocardium, as opposed to an intravenous injection. The venous return from the right coronary and circumflex arterial beds is undisturbed, so that BMS-181321 recirculates immediately through the right heart, pulmonary system, and left heart into the systemic circulation. The LAD vein was cannulated to allow the sampling of venous effluent for labeled metabolic byproducts; the venous cannula drained into the pulmonary cavity and returned blood to the venous system through the femoral vein by a roller pump. The advantages of this preparation are the capability to easily control coronary blood flow in the three circulations, selective administration of radiolabeled tracers, and accurate determination of the amount of tracer presented to the heart. The injection of tracer via the intracoronary route, nonetheless, means a greater presentation of the initial bolus of tracer than would be expected with intravenous injection. With recirculation of venous blood, the subsequent arterial concentration would be similar to an intravenous injection. Blood levels of BMS-181321 after injection were not measured in the present study to confirm the changes after recirculation.
The presence of myocardial ischemia in the perfusion bed of the LAD artery was confirmed by the decline in systolic wall thickening. Furthermore, metabolic changes of myocardial ischemia were also present, with a decrease in myocardial oxygen consumption and an increase in glucose utilization. Fatty acid utilization also declined. The recovery in systolic function with reperfusion was incomplete, suggestive of stunning after the ischemic insult. Furthermore, previous histological studies in swine subjected to longer period of ischemia, 60% flow reduction for 1 hour, demonstrated patchy areas of subendocardial infarction only.12 Thus, the ischemic insult was not of sufficient magnitude for precipitation of transmural infarction.
Myocardial uptake of BMS-181321 was compared with regional blood flow during ischemia. In the normally perfused myocardium of the circumflex and right coronary arteries, uptake of BMS-181321 was uniform. In the LAD perfusion bed, however, a significant negative correlation of BMS-181321 activity existed compared with myocardial blood flow. The negative correlation suggests that the uptake was in proportion to the degree of ischemia in the myocardium. The dependence of BMS-181321 uptake on tissue oxygen level is further highlighted by the decrease in the epicardial/endocardial ratio of BMS-181321 activity associated with the expected increase in the epicardial/endocardial ratio of blood flow during ischemia. The relation between ischemia and retention of BMS-181321 is similar to those for fluoromisonidazole1 in the heart and for BMS-181321 in the brain.4
BMS-181321 represents a novel nitroimidazole with a propylene amine oxime group added to form a coordination complex with 99mTc.2 The class of nitroimidazole compounds has been found to have diverse pharmacotherapeutic benefits of antimicrobial and antineoplastic activity.13 Cellular uptake of nitroimidazoles occurs by passive uptake; lipophilicity of BMS-181321 is comparable to that of other neutral species, such as 99mTc-propylene amine oxime and 99mTc-l,l,-ethyl cysteinate dimer.14 Within the cell, reduction of the nitro group occurs by nitroreductase enzymes in the cytoplasm.15 In normoxic conditions, oxygen acts as an electron acceptor, and the original nitroimidazole is re-formed; in the setting of hypoxia, the decrease in cellular oxygen levels retards the oxidative process, allowing the reduced compound to undergo further reduction.16 17 With reduction of the nitroimidazole to hydroxylamine and amine derivatives, cellular retention of the compound occurs either because of the decrease in efflux of the more hydrophilic intermediates or covalent binding of the intermediates to intracellular proteins.16 18 In the setting of more constant plasma levels with repetitive oral administration of a nitroimidazole, the reduced nitroimidazole lessens the efflux of the unreduced compound, favoring further reduction of the nitroimidazole.19 In the setting of bolus administration of a nitroimidazole, regional uptake will be related to blood flow and the first-pass extraction of tracer. With the clearance of tracer from the blood pool, equilibrium conditions will not apply, lessening the impact of intracellular binding of tracer on net cellular retention. This is particularly true in the extracorporeally perfused working swine heart preparation, in which the recirculation of intact tracer to the heart is limited by passage through the arterial circulation first. Regional differences in nitroimidazole will be most dependent on the lack of oxidation of reduced tracer either in the setting of bolus administration or in more constant plasma levels. Greater levels of oxygen deprivation during hypoxia or ischemia would be expected to be associated with greater retention of the nitroimidazole. Although we did not directly measure tissue oxygen levels, the increase in endocardial BMS-181321 activity in the setting of decreased endocardial perfusion with ischemia supports the hypothesis that greater retention of BMS-181321 will occur in regions of more marked oxygen deprivation.
In both normally perfused and ischemic/reperfused myocardium, a biexponential washout of BMS-181321 was seen; prolongation of the slow washout component in the ischemic tissue occurred, despite a return to normal baseline perfusion. The biexponential washout is similar to the kinetics of iodovinylmisonidazole.20 The rapidity of washout of the first component and its relative size suggest that the first component of BMS-181321 washout represents the clearance of free tracer. The change in washout rate of the slow component suggests that the second phase of washout represents the clearance of intracellular tracer, which has been reduced beyond the initial reduction step.15 16 The change in relative size of the first and second components also suggests that the first component represents not only clearance of free tracer from the extracellular space but also clearance of free intracellular tracer.
Critical to the clinical application of hypoxic markers is the differentiation of washout of the marker in ischemic versus normally perfused tissue.21 This time frame may be more rapid for BMS-181321 than for [18F]fluoromisonidazole, in which only 3 of 14 dogs had enough differentiation of ischemic and normally perfused myocardium for imaging at 45 minutes,1 compared with all 7 animals having uptake of BMS-181321 discernible in risk regions with ex vivo imaging after 70 minutes of reperfusion. However, since this difference may in fact be related to differences in imaging procedure or degree of ischemia between the two experimental preparations, further studies are necessary to directly compare the kinetics of [18F]fluoromisonidazole and BMS-181321.
The experimental preparation allowed for the intracoronary administration of BMS-181321. Calculation of the fraction of the total dose retained by the myocardium was possible by use of the gamma counter efficiency. The mean percent retention of BMS-181321 for normally perfused myocardium at 70 minutes of reperfusion was 0.005% of the injected dose per gram of tissue. Considering the intracoronary injection of BMS-181321, this fractional retention of the nitroimidazole is probably lower than the iodinated misonidazole, which had a retention of 0.003% of the intravenously injected dose in the normal heart.19 The maximal retention in the ischemic bed was twofold greater than in the normal bed. Thus, the more rapid washout of BMS-181321, the low retention of tracer in normal myocardium, and the twofold increase in maximal activity in the ischemic bed may assist in the differentiation of ischemic and normal myocardial regions. The low myocardial retention of BMS-181321 and high hepatic activity raise concern about imaging of ischemic myocardium in vivo. The experimental preparation with delivery of BMS-181321 to the coronary circulation before the systemic arterial or venous circulation would mean a higher coronary arterial concentration of tracer than would be expected from an intravenous injection. Thus, the heart-to-liver ratio of 0.58 may be lower in studies that used an intravenous injection of the tracer. Dynamic imaging in this study was also performed with an open-chest preparation, which hindered extrapolation to in vivo imaging with the chest intact and with intravenous injection of tracer. However, in vivo studies have been performed in a cat model of stroke using single photon emission computed tomography (SPECT) brain imaging and a intact dog preparation with myocardial ischemia with cardiac SPECT imaging, demonstrating the imaging capabilities of BMS-181321 of cerebral4 and myocardial22 ischemia in vivo. These initial cerebral and cardiac SPECT studies suggest that further studies of imaging of myocardial ischemia in a closed-chest preparation are warranted.
The classic method of imaging myocardial ischemia with radionuclide tracers has been dependent on demonstration of differences in myocardial perfusion or tracer kinetics. These methods, such as 201Tl stress/redistribution or serial 99mTc-sestamibi imaging, require two image sets to visualize myocardial ischemia as a reversible region of tracer deficit. Thus, myocardial ischemia has been characterized as a negative image by conventional tracer kinetics. Interest has been present in imaging myocardial ischemia with [18F]fluoromisonidazole, since its tracer kinetics may alter the differentiation of ischemia from normal or infarcted myocardium as one single positive image. The use of 99mTc as the radionuclide for BMS-181321 eases some of the synthesis and imaging constraints present with [18F]fluoromisonidazole with the availability of molybdenum generators and standard gamma cameras. Because of its potential availability, there are several clinical scenarios in which BMS-181321 may be applied to positively image ischemic myocardium. Those settings include acute myocardial infarction with failed thrombolysis or exercise-associated ischemia with critical coronary artery disease. Further in vivo studies are necessary to continue the clinical feasibility of imaging myocardial ischemia positively.
This research was supported in part by grants from Bristol-Myers Squibb, Princeton, NJ, and from the National Institutes of Health (Heart, Lung, and Blood Institute, FIRST Award HL-47003 to Dr Stone), the Rennebohm Foundation of Wisconsin, and the Oscar Mayer Research Fund. The expert technical support from Larry F. Whitesell, Catherine R. Kidd, Gregory R. Thomas, Daniel K. Paulson, Emmanuel Scarbrough, and Alice M. Eggleston of the Cardiovascular Research Laboratory is gratefully acknowledged. The secretarial assistance of Thankful Sanftleben in preparation of the manuscript is also appreciated.
Dr Nunn, Mr Kuczynski, and Dr Rumsey were employed by Bristol-Myers Squibb, which also supported this study with a research grant. They are now employed by Bracco Research USA. Dr Nickles is the recipient of another research development grant from Bristol-Myers Squibb.
Presented in part at the 40th Annual Meeting of the Society of Nuclear Medicine, Toronto, June 8-11, 1993.
- Received August 3, 1994.
- Revision received February 7, 1995.
- Accepted February 25, 1995.
- Copyright © 1995 by American Heart Association
Shelton ME, Dence CS, Hwang D-R, Herrero P, Welch MT, Bergmann SR. In vivo delineation of myocardial hypoxia during coronary occlusion using fluorine-18 fluoromisonidazole and positron emission tomography: a potential approach for identification of jeopardized myocardium. J Am Coll Cardiol. 1990;16:477-485.
Liedtke, AJ, DeMaison L, Eggleston AE, Cohen LM, Nellis SH. Changes in substrate metabolism and effects of excess fatty acids in reperfused myocardium. Circ Res. 1988;62:535-542.
Renstrom B, Nellis SH, Liedtke AJ. Metabolic oxidation of glucose during early myocardial reperfusion. Circ Res. 1989;65:1094-1101.
Renstrom B, Nellis SH, Liedtke AJ. Metabolic oxidation of pyruvate and lactate during early myocardial reperfusion. Circ Res. 1990;66:282-288.
Nellis SH, Liedtke AJ, Renstrom B. Distribution of carbon flux within fatty acid utilization during myocardial ischemia and reperfusion. Circ Res. 191;69;779-790.
Liedtke AJ, Renstrom B, Nellis SH. Correlation between [5-3H]glucose and [U-14C]deoxyglucose as markers of glycolysis in reperfused myocardium. Circ Res. 1992;71:689-700.
Mulnix TL, Nickles RJ, Stone CK. A CAMAC-modular gamma camera for 50<E<511 keV research imaging. J Nucl Med. 1994;35:178P. Abstract.
Winer BJ. Statistical Principles in Experimental Design. New York, NY: McGraw-Hill; 1971:201-204.
Liedtke AJ, Renstrom B, Nellis SH, Subramanian R, Woldegiorgis G. Myocardial metabolism in chronic reperfusion after nontransmural infarction in pig hearts. Am J Physiol. 1993;265:H1614-H1622.
Edwards DI. Nitroimidazole drugs: action and resistance mechanisms, I: mechanism of action. J Antimicrob Chemother. 1993;31:9-20. Review.
Pirro JP, Di Rocco RJ, Narra RK, Nunn AD. Relationship between in vitro transendothelial permeability and in vivo single-pass brain extraction. J Nucl Med. 1994;35:1514-1519.
Van Os-Corby D, Koch DJ, Chapman JD. Is misonidazole binding to mouse tissues a measure of cellular pO2? Br J Cancer. 1987;56:3487-3494.
Chan Y-W, Linder KE, Jayetilah PG, Patel BC, Nunn AD, Romero NR, Ramsey WC. In vitro studies on the hypoxic retention of a novel technetium-99m labeled nitroimidazole in rat hearts. J Nucl Med. 1994;35:19P. Abstract.
Martin GV, Cerqueira MD, Caldwell JH, Rasey JS, Embree L, Krohn KA. Fluoromisonidazole: a metabolic marker of myocyte hypoxia. Circ Res. 1990;67:240-244.
Martin GV, Biskupiak JE, Caldwell JH, Rasey JS, Krohn KA. Characterization of iodovinylmisonidazole as a marker for myocardial hypoxia. J Nucl Med. 1993;34:918-924.
Nunn AD. Is nuclear medicine viable and can it measure viability? J Nucl Med. 1993;34:924-926.
Rumsey WL, Kucynski B, Patel B, Bauer A, Narra RK, Eaton SM, Nunn AD, Strauss HW. SPECT imaging of ischemic myocardium using a technetium 99m-nitroimidazole ligand. J Nucl Med. (in press).