The frequency of false positive results obtained from the inferior myocardial region using single photon emission computed tomography (SPECT) myocardial perfusion scintigraphy is significantly higher than that obtained from other regions. Several methods, such as prone-position imaging, have been proposed to overcome this diagnostic problem. The aim of the present study was to compare the results of Tc-99m-sestamibi gated SPECT and Tl-201 prone SPECT in the differentiation of inferior wall artifacts from true defects. For this purpose, 38 subjects, whose coronary anatomies were documented on angiography, underwent same-day stress-rest Tc-99m-sestamibi gated SPECT and Tl-201 stress-reinjection-prone (whose standard supine images demonstrated fixed defects on the inferior wall) SPECT. Gated SPECT was performed by 8 frames per cycle acquisition over a 180 degrees rotation on 30 projections. Four gated SPECT slices were obtained on mid-ventricular vertical long axis, horizontal long axis and apical and basal short axis planes, and displayed in cine-format. Both Tl-201 prone imaging and Tc-99m-sestamibi gated analysis increased the specificity of inferior wall disease detection remarkably from 54% to 85% and 46% to 82%, respectively (P<0.05). The difference between diagnostic accuracies was not significant (80% and 82%, respectively) (P> 0.05). The positive predictive values for true defects were 96% for Tl-201 prone imaging and 94% for Tc-99m-sestamibi gated imaging. Based on segmental analysis, the two modalities showed fair agreement (kappa = 0.44 for standard supine protocols, kappa = 0.46 for Tl-201 prone and Tc-99m-sestamibi gated SPECT). It can be concluded that Tc-99m-sestamibi gated SPECT, requiring only two-step acquisition, may potentially increase the test specificity for coronary artery disease (CAD) of the inferior wall as well as does Tl-201 sh ess-reinjection-prone SPECT. By giving functional information, it seems the most practical method in daily use for supplying the most extensive information about patients with suspected or known CAD. ((C) 2000 Lippincott Williams & Wilkins).