Effects of altered left ventricular geometry on quantitative technetium 99m sestamibi defect size in humans: perfusion imaging during coronary angioplasty.

BACKGROUND: Serial myocardial perfusion imaging is used to assess exercise-induced myocardial ischemia and myocardial risk area, salvage, and viability in patients with myocardial infarction. In an experimental animal model it has been shown that abnormal regional wall motion and altered left ventricular geometry can produce apparent perfusion defects independent of changes in blood flow. The effects of regional alteration in ventricular geometry on perfusion images in humans are not defined. The purpose of our investigation was to evaluate quantitatively the effect of altered left ventricular geometry on myocardial perfusion imaging with technetium 99m sestamibi during coronary angioplasty. METHODS AND
RESULTS: Nine patients with normal baseline left ventricular function referred for angioplasty of the left anterior descending coronary artery were studied. 99mTc sestamibi was administered intravenously before angioplasty. Baseline planar electrocardiographic-gated imaging was performed. Imaging was repeated in the catheterization laboratory during angioplasty vessel occlusion when altered left ventricular geometry was produced and again later after angioplasty. Summed static, end-systolic, and end-diastolic images were generated from the electrocardiographic-gated acquisitions. Circumferential count profiles of images obtained during percutaneous transluminal coronary angioplasty (PTCA) were compared with those of a normal 99mTc sestamibi database and their own baseline images. Defect integral (the area below the reference profile) and nadir (maximum percent decrease in activity) were derived. Compared with a normal database, new quantitative defects appeared on PTCA-summed images in only two patients. The defects were small to moderate in size. However, compared with their own baseline profile, six patients had quantitative defects during PTCA (mean defect integral 3 +/- 2; mean defect nadir 12% +/- 7%). Defect nadir was larger on end-diastolic images compared with summed images (22% +/- 7% and 12% +/- 7%, respectively; p < 0.05).
CONCLUSIONS: Altered left ventricular geometry may create apparent, albeit small, planar myocardial perfusion defects in humans. Changes in defect size on serial images may be only partially caused by changes in regional wall motion or geometry.