Abstract
3343
Introduction: Myocardial infarction (MI) often results in infarct thinning and expansion and left ventricular (LV) remodeling and can ultimately lead to heart failure (HF) and increased mortality. The induction of proteolytic enzymes in the myocardium, including the matrix metalloproteinases (MMPs) occurs early in patients following MI and is related to the degree of adverse LV remodeling. In evaluation of post-MI remodeling, we have previously established an imaging approach for assessment of regional MMP activation within the LV myocardium utilizing single-photon emission computed tomography (SPECT) imaging of a 99mTc-labeled MMP targeted radiotracer, 99mTc-RP805. The transition from traditional SPECT cameras with Na-Iodide detectors, photomultipliers, and rotating gantries to higher sensitivity and higher resolution solid state cadmium-zinc-telluride (CZT) detector systems enables creation of high resolution cardiac gated images from dynamic listmode datasets. There is a clear advantage in the clinical setting in the ability to assess LV function with early SPECT images and late “hotspot” images of MMP activation without the need for injection of other radiotracers or iodinated contrast CT. The purpose of this study was to evaluate early (0-5 min) 99mTc-RP805 cardiac gated images for quantitative assessment of blood pool images for determination of left ventricular function in relation to late (4 hr) targeted “hotspot” images for quantification of LV MMP activity following an MI.
Methods: Dynamic 99mTc-RP805 images were retrospectively reconstructed for the initial 5 min of imaging in swine (n=5) 3 days following creation of reperfused infarction by 90-minute occlusion of the left anterior descending coronary artery. Pigs had dynamic imaging over 4 hours to assess late myocardial uptake of 99mTc-RP805. In each animal following the ECG-gated 99mTc-RP805 SPECT imaging, we performed a high resolution (0.625 mm) ECG-gated contrast CT which was used as the gold standard for assessment of cardiac function. A delayed enhancement CT was used for identification and co-localization of radiotracer activity in the MI region. The ECG-gated 99mTc-RP805 SPECT scans were analyzed with commercial software (4DM, Invia) and compared with ECG-gated contrast CT images (AW software, GE Healthcare) for determination of left ventricle ejection fraction (LVEF).
Results: Figure 1 shows the excellent image quality obtained in representative initial ECG-gated 99mTc-RP805 images of blood pool activity and image analysis. There was a high correlation between the LVEF determined from ECG-gated contrast CT and the early ECG-gated 99mTc-RP805 SPECT blood pool with an R2 of 0.9925 (Figure 1C). A Bland-Altman analysis shows excellent agreement of the data and a low bias (Figure 1D). A representative time activity curve displays the relative uptake in the global LV myocardium vs blood input function (Figure 1E). Focal uptake in the infarct area is clearly defined at 4 hr post-MI (Figure 1F).
Conclusions: While regional and global LV function can be derived from ECG-gated SPECT perfusion images, this is not possible with “hot spot” targeted MMP imaging using 99mTc-RP805. However, ECG-gated 99mTc-RP805 blood pool imaging provides a means to determine left ventricle function post-MI using initial blood pool images without additional radiotracers. Thus, important and clinically relevant data can be derived from initial bloodpool activity of a targeted “hotspot” imaging agent and correlated with focal radiotracer uptake in the myocardial infarct region.