Search

A central assumption in SPECT is that the projection data are "consistent," that is, the camera views an unchanging distribution during acquisition. Several new radiotracers of interest, including 99mTc-teboroxime (Cardiotec), have rapid clearance from the myocardium. Furthermore, the washout is different in normal and ischemic tissues. We used computer simulations to estimate the effect of this differential washout on quantification of the severity of ischemia. We simulated defect-to-normal myocardial activity ratios of 1 (no defect), 0.8, 0.6, 0.4, 0.2, and 0 (complete defect), with single defects placed either in the lateral wall or apex, and SPECT acquisitions of 1, 3, 6, 12, and 24 total minutes. We modeled washout with a monoexponential curve whose clearance half-time was 5.9 min for "normal myocardium" and 9.3 min for "ischemic myocardium." We found that differential washout from normal and ischemic zones produced image artifacts and errors in defect quantification for acquisitions longer than 3 min. With longer acquisitions, the degree of ischemia was significantly underestimated, with increasing error at longer acquisition times. In addition, in the "no defect" situation an apparent small lateral wall defect (relative to the apex) was present. Finally, lateral wall defects produced artifacts (streaks and reduced apparent activity) in the opposite (medial) wall. Differential normal/ischemic zone washout during SPECT acquisition produces artifacts and errors in quantification, whose severity is dependent on acquisition length, actual defect severity, and defect location.

This study was done to determine whether the rapidly clearing myocardial perfusion agent 99mTc-teboroxime (SQ 30217, Cardiotec) could be combined with tomographic imaging to accurately quantify regional myocardial blood flow distribution in anesthetized dogs. Following stenosis of the anterior descending (LAD, n = 10) or circumflex (LCX, n = 5) coronary arteries, teboroxime was administered simultaneously with radioactive microspheres, at rest and following infusion of dipyridamole (0.15 mg/kg/min x 4 min). Tomographic imaging began 1 min after each teboroxime injection and continued for 12 min. For LAD stenosis, when the dipyridamole study was performed first, teboroxime activity in the center of the ischemic region was closely correlated with tissue microsphere content. However, the severity of the dipyridamole-induced flow deficit was underestimated by teboroxime when the rest study was performed first. Our results show that despite rapid myocardial clearance, tomographic imaging of 99mTc-teboroxime provides reasonably accurate quantitation of dipyridamole-induced anterior wall perfusion defects, but that the flow deficit is underestimated when a rest study is performed first or when the defect is located in the inferior wall.

Quantitation of myocardial perfusion with thallium-201 (201Tl) SPECT is limited by finite resolution and image noise. This study examined whether Wiener filtering could improve quantitation of the severity of myocardial perfusion deficits. In 19 anesthetized dogs, adjustable stenoses were placed on the left anterior descending (LAD, n = 12) or circumflex (LCx, n = 7) arteries. Thallium-201 SPECT images were acquired during maximal coronary vasodilation with dipyridamole, and simultaneous measurements of myocardial blood flow were made with microspheres. The relationship between SPECT and microsphere flow deficits in the LAD region was significantly better (p less than 0.05) with Wiener filtering (Y = 0.90X + 0.03, r = 0.78) than with conventional Hanning filtering (Y = 0.66X + 0.34, r = 0.61). Similarly, in the LCx region the relationship between SPECT and microsphere perfusion deficits was better (p less than 0.01) with the Wiener filter (Y = 0.91X + 0.07, r = 0.66) than with the Hanning filter (Y = 0.36X + 0.50, r = 0.40). Wiener filtering improves quantitation of the severity of regional myocardial perfusion deficits, allowing better assessment of the functional significance of coronary artery stenoses.

To determine whether technetium-99m-hexakis-2-methoxyisobutyl isonitrile (SESTAMIBI) remains fixed in the myocardium following its initial uptake or undergoes time-related redistribution, anesthetized dogs underwent occlusion of the anterior descending coronary artery for 6 min, followed by 3-hr reperfusion. Technetium-99m-SESTAMIBI and thallium-201 (201Tl) were injected intravenously after 1 min occlusion and regional myocardial blood flow was measured with radioactive microspheres. Tomographic imaging of Tc-SESTAMIBI revealed a perfusion defect with slight but definite filling in over 2 hr. Quantitative analysis indicated a significant rise in the nadir and decrease in the width of the defect in circumferential profile curves. After 3-hr of reperfusion, Tc-SESTAMIBI activity in the previously ischemic area was always greater than the activity of microspheres injected during coronary occlusion (mean normalized values, 0.32 versus 0.11, p less than 0.0001). Our results indicate that following transient ischemia and reperfusion, Tc-SESTAMIBI clearly undergoes myocardial redistribution, although more slowly and less completely than 201Tl.

This study was done to determine whether [99mTc]methoxyisobutyl isonitrile (RP30), a nonredistributing myocardial perfusion agent, could be used to quantify regional myocardial blood flow distribution during ischemia and reperfusion, employing sequential injections of tracer, tomographic imaging, and appropriate image subtraction. Dogs underwent transient (6 min) coronary artery occlusion, followed by two paired injections of RP30 and radioactive microspheres combined with tomographic imaging, the first during coronary occlusion and the second after 60 min of reperfusion. To obtain a true image representative of reperfusion, the first set of images was corrected for 99mTc decay and subtracted from the second. During occlusion, tissue microsphere content and scintigraphic RP30 activity in the center of the ischemic region (both expressed as a fraction of the nonischemic region) were closely correlated, although RP30 consistently exceeded microsphere content (0.43 +/- 0.03 vs. 0.24 +/- 0.04, p less than 0.01). Direct tissue counting of RP30 confirmed its relative excess in ischemic myocardium. Reperfusion was successful in 7/8 dogs, with an increase in RP30 activity to 0.98 +/- 0.04 compared to 0.89 +/- 0.03 for microspheres (p = N.S.). In one dog with microsphere-documented persistent ischemia, the RP30 defect was still present after reflow. Our results indicate that because of the lack of myocardial clearance and redistribution, repeat injections of RP30 can be used to quantify serial changes in regional myocardial blood flow.

We describe a technique for correction of artifacts in exercise 201Tl single photon emission computed tomography (SPECT) images arising from abrupt or gradual translational movement of the heart during acquisition. The procedure involves the tracking of the "center of the heart" in serial projection images using an algorithm which we call "diverging squares". Each projection image is then realigned in the x-y plane so that the heart center conforms to the projected position of a fixed point in space. The shifted projections are reconstructed using the normal filtered backprojection algorithm. In validation studies, the motion correction procedure successfully eliminated movement artifacts in a heart phantom. Image quality was also improved in over one-half of 36 exercise thallium patient studies. The corrected images had smoother and more continuous left ventricular walls, greater clarity of the left ventricular cavity, and reduced streak artifacts. Rest injected or redistribution images, however, were often made worse, due to reduced heart to liver activity ratios and poor tracking of the heart center. Analysis of curves of heart position versus projection angle suggests that translation of the heart is common during imaging after exercise, and results from both abrupt patient movements, and a gradual upward shift of the heart. Our motion correction technique appears to represent a promising new approach for elimination of movement artifacts and enhancement of resolution in exercise 201Tl cardiac SPECT images.

The quantitative relationship between fractional myocardial thallium uptake and radioactive microsphere-determined flow was studied in 33 open chest dogs under baseline conditions during increased coronary flow (dipyridamole), decreased coronary flow (propranolol and coronary artery stenosis), inhibition of Na-K ATPase (ouabain), and regional infarction. Myocardial contents of thallium and microspheres were compared in left ventricular (LV) biopsies taken 5, 10, 15, 30, and 60 min after thallium injection, expressed as fractions of injected dose. Maximal LV thallium uptake occurred 10 min after injection and the 10-min values were therefore used for subsequent comparisons. Combining all dogs, fractional LV thallium content (% injected dose) correlated well with fractional LV blood flow (% cardiac output) (r = 0.95). However, for fractional LV flows in the low, normal, or moderately elevated range (LV flow/cardiac output less than 9%), thallium content consistently exceeded flow by about 15%. This relationship was not altered by ouabain or regional ischemia or infarction. For greatly elevated fractional LV flows (greater than 9%), thallium content was not significantly different from flow. To explain these differences, myocardial and systemic extraction fractions for thallium were determined in eight dogs with a dual tracer method. At baseline, myocardial extraction fraction was significantly greater than systemic (88 +/- 0.4% compared with 75 +/- 1%, p less than 0.001). During dipyridamole, myocardial extraction fraction decreased and myocardial and systemic values were no longer significantly different (82 +/- 1% compared with 79 +/- 1%). These results show that the fraction of injected thallium dose taken up by the LV myocardium exceeds the delivered fraction of cardiac output over a wide range of LV flows, and is not altered by ouabain-induced inhibition of sodium-potassium ATPase or regional myocardial infarction. This difference is explained by a greater myocardial than systemic extraction fraction for thallium. During high LV flows produced by dipyridamole, fractional LV thallium uptake and flow become similar as myocardial and systemic extraction fractions equalize.