In vitro validation of a simple tomographic technique for estimation of percentage myocardium at risk using methoxyisobutyl isonitrile technetium 99m (sestamibi)

Abstract

With the advent of technetium 99m-labeled myocardial blood flow agents, there is a need for a simple technique for quantitation of infarcted or jeopardized myocardium (IM). This study provides an in vitro validation of a simple technique based upon the analysis of three short-axis slices through the heart following emission computed tomography. All acquisitions were performed using a static cardiac phantom containing pertechnetate Tc 99 m. Activity in the phantom was adjusted so that the count density and myocardial-to-background ratio were comparable to those observed in patients. Plastic insets (range of sizes = 4%–72% of myocardium) were used to simulate transmural infarctions. Eighteen studies were acquired, each over 180° into a 64 × 64 matrix. Data were reconstructed using a Ramp Hanning filter with cut off at 0.7 times the Nyquist frequency. Short-axis slices of the myocardium were then generated, and representative apical (A), mid-ventricular (MV), and basal (B) slices were selected. For each slice, a circumferential profile was generated, and the average radius (R) was measured. The fraction (F) of the profile falling below a threshold value was considered to represent IM. Total IM was given by %IM =100 x (R _B F _B +R _MV F _MV + 0.67R _A F _A)/(R _B +R _MV + 0.67R _A), where the subscripts to R and F refer to the relevant short-axis slices. For a threshold set at 60% of peak, measured IM agreed closely with true IM (R ²=0.98, measured IM =1.01 x true IM −1.35). Measurement of % IM was not distorted by variations in slice radius or in slice selection. Maximum error in % IM occurred with a change in location of the infarct (approximately 4% for opposing walls). This technique permits rapid and accurate assessment of % IM with^99mTc-labeled myocardial blood flow agents.