Does quantification of myocardial perfusion SPECT study differ while image reconstruction is carried out using iteration algorithm instead of filtered back-projection?--preliminary report.

Abstract

BACKGROUND: The purpose of this study was to compare the performance of two reconstruction algorithms: conventional filtered back-projection (FBP) and an iterative algorithm--ITW--in quantitative analysis of myocardial perfusion SPECT studies. The defect size and defect severity were assessed on (99m)Tc-MIBI images reconstructed using both methods and estimation of sensitivity in the detection of perfusion deficits and myocardial viability were performed as well. METHODS AND RESULTS: The study group comprised 43 patients (38 men and 5 women) in the age of 40-73 years (mean 59 years). Heart perfusion scintigraphy was performed following an injection of 22 to 25 mCi 99mTc-MIBI for exercise and rest myocardial perfusion study. Images were reconstructed using FBP and ITW algorithms. Defect size (DS) was quantified by a threshold program and CEqual programme. Defect severity (nadir) was calculated as the ratio of minimal/maximum counts from bull?s eye polar map. Coronary arteriography was performed in all patients. RESULTS: Defect size calculated by threshold method on resting images did not differ between reconstruction methods (p=0.61 for cut-off 50% and p = 0.24 for cut-off 60%); defect severity was higher on images reconstructed with ITW (CI(0.95) = 2.4%;5.2% of maximal counts). CONCLUSIONS: Sensitivity for detection of heart perfusion defects and estimation of myocardial viability were similar on images reconstructed by both algorithms.