The purpose of this study was to assess the ability of quantitative Tl-201 tomography to identify and localize coronary artery disease (CAD). The study population consisted of 4l. patients (31 males, 10 females; mean age 55 +- 7 yr) including 14 with prior myocardial infarction who underwent both exercise Tl-201 myocardium SPECT and coronary angiography for the evaluation of chest pain. From the short axis and vertical long axis tomograms, stress extent polar maps were generated by Cedars-Sinai Medical Center prograrn, and the 9 stress defect extent (SDE) was quantified for each coronary artery territory. For the purpose of this study, the coronary circulation was divided into 6 arterial segments, and the myocardial ischemic score (MIS) was calculated from the coronary angiogram. Sensitivity for the detection of CAD (>50% coronary stenosis by angiography) by stress eXtent polar map was 95% in single vessel disease, and 100% in double and triple vessel deseases. Overall sensitivity was 97%<. Sensitivity and specificity for the detection of individual diseased vessels werc, respectively, 87% and 90% for the left anterior descending artery (LAD), 36% and 93% for the left circumflex artery (LCX), and 71% and 70%, for the right coronary artery (RCA). Concordance for the detection of individual diseased vessels between the coronary angiography and stress polar map was fair for the LAD (kappa=0.70), and RCA (kappa=0.41) lesions, whereas it was poor for the LCK lesions (kappa =0.32) There were siginificant correlations between the MIS and SDE in LAD (rs=0. 56, p=0.0027), and RCA territory (rs=0.60, p=0.0094). No significant correlation was found in LCX territory. When total vascular territories were combined, there was a significant correlation between the MIS and SDE (rs=0.42, p=0,0116). In conclusion, the quantitative analysis of TI-201 tomograms appears to be accurate for determining the presence and location of CAD.