The use of dual-head gamma camera modified positron coincidence detection (PCD) is a new, alternative method of 2-[18F]fluoro-2-deoxy-D-glucose (FDG) imaging. This study investigated the potential ability of evaluating myocardial viability in patients with ischaemic heart disease by FDG imaging using PCD. A total of 21 patients (18 male, three female; mean age 59.7+/-8.5 years) with a history of previous myocardial infarction and confirmed coronary angiography underwent FDG PCD and FDG PET after oral glucose loading (75 g). Quantitative analysis was compared between images of FDG PCD and FDG PET. A significant linear correlation between the segmental percentage of FDG uptake obtained by PCD and PET was observed (r=0.63, P<0.001). By receiver operating characteristic (ROC) analysis, using FDG PET as the 'gold standard', at the 50% threshold value in PET, FDG PCD showed a sensitivity of 92% and specificity of 63% in detecting myocardial viability. Regional analysis showed lower agreement of FDG PCD and FDG PET in the inferior (79%) and septal (70%) walls compared with the other walls. Quantitative evaluation of myocardial viability using FDG PCD yielded comparable clinical results in apex, anterior and lateral walls to that of FDG PET. However, the agreement was lower in the inferior and septal walls. Therefore, results of FDG PCD should be carefully interpreted in evaluating myocardial viability in the inferior and septal walls. The application of a measured attenuation correction and scatter correction are needed to improve the detectability of myocardial viability in FDG imaging by coincidence gamma camera.