BACKGROUND: Serial perfusion imaging with 99mTc-labeled sestamibi has been useful in the assessment of myocardial salvage from reperfusion therapy during acute myocardial infarction. Studies in animal models have shown that discernible perfusion defects can be created by left ventricular asynergy from partial volume effects in the setting of homogenous perfusion tracer distribution. The purpose of this study was to examine the influence of gating on serial perfusion images during acute myocardial infarction to determine the magnitude of potential partial volume effects. METHODS AND RESULTS: 99mTc-labeled sestamibi was injected into 18 patients during acute myocardial infarction and 29 patients 5 to 8 days after myocardial infarction. Tomographic imaging was acquired in gated format (16 frames per R-R cycle of the electrocardiogram) for each set of images. All frames were summed to produce ungated images. Tomographic images were quantified on three different thresholds to define the perfusion defect: 50%, 60%, and 70% of maximal counts. Severity of perfusion defects was calculated as the lowest ratio of minimum/maximum counts on five short-axis slices. Regional wall motion was assessed subjectively on the gated images by cine-loop display. Radionuclide ventriculography was performed at 6 weeks. There was a close correlation between perfusion defect size on ungated images and end-diastolic and end-systolic images independent of the quantitative threshold used (r = 0.90 to 0.93; p < 0.0001 for all correlations). Gated images provided consistently significantly greater estimates of perfusion defect size and severity by a small increment (3% to 9% of the left ventricle; p < 0.05 for all comparisons) independently of the quantitative threshold used or the time of imaging (acute or late). Ungated images provided slightly better correlations with left ventricular ejection fraction at 6 weeks independently of the quantitative threshold used and despite significant wall motion abnormalities present on both the acute and final studies. CONCLUSIONS: The differences between perfusion defect size for gated and ungated images were highly significant as a group, but the magnitude of difference was small and not clinically relevant. The larger estimates provided by end-diastolic gated images are opposite the difference expected if partial volume effects were significantly influencing perfusion defect size. Partial volume effects appear to have minimal impact on clinical tomographic imaging during acute myocardial infarction for the quantification of myocardium at risk and infarct size.
BACKGROUND: Serial perfusion imaging with 99mTc-labeled sestamibi has been useful in the assessment of myocardial salvage from reperfusion therapy during acute myocardial infarction. Studies in animal models have shown that discernible perfusion defects can be created by left ventricular asynergy from partial volume effects in the setting of homogenous perfusion tracer distribution. The purpose of this study was to examine the influence of gating on serial perfusion images during acute myocardial infarction to determine the magnitude of potential partial volume effects. METHODS AND RESULTS: 99mTc-labeled sestamibi was injected into 18 patients during acute myocardial infarction and 29 patients 5 to 8 days after myocardial infarction. Tomographic imaging was acquired in gated format (16 frames per R-R cycle of the electrocardiogram) for each set of images. All frames were summed to produce ungated images. Tomographic images were quantified on three different thresholds to define the perfusion defect: 50%, 60%, and 70% of maximal counts. Severity of perfusion defects was calculated as the lowest ratio of minimum/maximum counts on five short-axis slices. Regional wall motion was assessed subjectively on the gated images by cine-loop display. Radionuclide ventriculography was performed at 6 weeks. There was a close correlation between perfusion defect size on ungated images and end-diastolic and end-systolic images independent of the quantitative threshold used (r = 0.90 to 0.93; p < 0.0001 for all correlations). Gated images provided consistently significantly greater estimates of perfusion defect size and severity by a small increment (3% to 9% of the left ventricle; p < 0.05 for all comparisons) independently of the quantitative threshold used or the time of imaging (acute or late). Ungated images provided slightly better correlations with left ventricular ejection fraction at 6 weeks independently of the quantitative threshold used and despite significant wall motion abnormalities present on both the acute and final studies. CONCLUSIONS: The differences between perfusion defect size for gated and ungated images were highly significant as a group, but the magnitude of difference was small and not clinically relevant. The larger estimates provided by end-diastolic gated images are opposite the difference expected if partial volume effects were significantly influencing perfusion defect size. Partial volume effects appear to have minimal impact on clinical tomographic imaging during acute myocardial infarction for the quantification of myocardium at risk and infarct size.