Au-195m is a radio-isotope with an ultra-short half-life with which multiple sequential evaluations of ventricular function can be made. In order to evaluate the reliability and reproducibility of analyses of overall and regional ventricular function by radio-isotope ventriculography with Au-195m we studied 10 healthy volunteers and 12 patients with coronary artery disease. Each subject underwent 4 first-pass studies: 1 with Tc-99m and, 10 minutes later, 3 with Au-195m (2 basal studies separated by 3-5 minutes interval and, 10 minutes later, 1 after s.l. nitroglycerin administration). Regional wall motion was analyzed and ejection fraction and peak count rate were determined in each test. Our study showed that the ejection fraction obtained with Au-195m was reproducible (r = 0.98) and correlated well with the ejection fraction determined by using Tc-99m (r = 0.98). The values of the peak count rate obtained with Tc-99m were higher than those obtained with Au-195m. Due to the specially designed collimator and the technical characteristics of the gamma-camera we used, we were able to record sufficiently high count-rates to evaluate regional wall motion, and this analysis was also found to be reliable and reproducible. After s.l. nitroglycerin administration, normal volunteers showed a significant increase of ejection fraction in comparison with basal acquisitions (p less than 0.05), while a wide range of responses was observed in the group of patients with coronary artery disease. We conclude that radio-isotope ventriculography with Au-195m is reliable and reproducible and could be a valid method of monitoring rapid variations induced in overall and regional left ventricular function.