The clinical diagnostic accuracy of 2-day stress/rest quantitative Technetium-99m (Tc-99m) methoxy-isobutyl-isonitrile (Tc-sestamibi) single photon emission computerized tomography (SPECT) was assessed in a validation population of 61 patients from two different sites using two different camera/computer systems. The study population was made up of 53 catheterized patients, 29 from Cedars-Sinai Medical Center (CSMC) and 24 from the University of Texas Southwestern Medical Center (UTSMC), and eight UTSMC patients with a less than 5% pre-test likelihood of coronary artery disease. Interpretation employed gender-specific normal limits developed in an additional 15 men and 8 women at CSMC with less than a 5% likelihood of significant coronary artery disease. The results from CSMC compared with those from UTSMC were not different from each other. The overall sensitivity for detection of patients with coronary artery disease (greater than or equal to 50% stenosis) was 94% (CSMC: 92%, UTSMC: 95%). Overall specificity in the five patients with normal coronary arteriograms was 80% (CSMC: 67%, UTSMC: 100%). The normalcy rate in patients with a low likelihood of coronary artery disease was 88%. Vessel sensitivity was 85% (CSMC: 84%, UTSMC: 85%), while vessel specificity was 71% (CSMC: 72%, UTSMC: 69%). There was also no significant difference in the sensitivities and specificities between male and female populations. In addition, the agreement with coronary angiography for assessment of disease extent (normal coronary arteriogram, single or multivessel disease) was 75% (kappa = 0.6 +/- 0.1). This study demonstrated that Tc-sestamibi SPECT by quantitative analysis is accurate for the detection and localization of coronary artery disease. Furthermore, the CSMC quantitative method was shown to provide similar diagnostic accuracy when applied to data acquired at a different site using a different camera/computer system.