MULTICENTER TRIAL VALIDATION FOR QUANTITATIVE-ANALYSIS OF SAME-DAY REST-STRESS TECHNETIUM-99M-SESTAMIBI MYOCARDIAL TOMOGRAMS

The accuracy of an automated quantitative analysis of same-day rest/st ress Tc-99m sestamibi SPECT images for detection and localization of c oronary artery disease (CAD) was assessed in a multicenter trial consi sting of 161 patients from 7 different clinical sites utilizing variou s camera computer systems. Methods: Of the 161 patients, 102 had angio graphically documented coronary artery disease, 22 had normal coronary arteriograms, and 37 had a low (< 5% likelihood of coronary artery di sease based on their age, sex, symptoms and the results of their exerc ise electrocardiograms. The patients were studied using previously opt imized image acquisition and processing protocols. An additional popul ation consisting of 45 patients with single-vessel disease were evalua ted to determine the optimal criteria for detection of CAD, Results: T he quantitative analysis method was associated with an overall sensiti vity of 87%, specificity of 36%, and normalcy rate (true negative rate in the tow likelihood patients) of 81%. Sensitivity for overall detec tion of disease was similar (90%) in patients with and without myocard ial infarction (90% versus 89%). The sensitivities and specificities f or identification of disease in individual coronary arteries were, res pectively, 69% and 76% for LAD, 70% and 80% for LCX, and 77% and 85% f or RCA. Conclusion: The results of this study demonstrate that the new objective quantitative method for analysis of same-day rest/stress Te -99m sestamibi SPECT images is accurate for detection and localization of CAD and correlates highly with expert visual interpretation.