We have developed new software which can evaluate left ventricular (LV) diastolic functional parameters from a quantitative gated SPET (QGS) program. To examine its accuracy, we compared these findings with the LV diastolic functional indices obtained from gated radionuclide ventriculography (RNV). Twenty-four patients were selected for this study. Gated SPET with technetium-99m tetrofosmin was performed and the QGS program was used with a temporal resolution of 32 frames per R-R interval. The LV volume of each frame was calculated and four harmonics of Fourier series were retained for the analysis of the LV volume curve. From this fitted curve and its first derivative curve, we derived LV systolic functional indices, e.g. ejection fraction (EF), peak ejection rate (PER) and time to PER (TPER), as well as LV diastolic functional variables, e.g. 1/3 filling fraction (1/3 FF), peak filling rate (PFR) and time to PFR (TPFR). Within 5+/-2 days, gated RNV was performed and diastolic functional parameters were determined by the same method. No significant difference was observed between the variables calculated by gated SPET and by gated RNV. There was a good correlation between EF, PER, TPER, 1/3 FF, PFR and TPFR determined by these two methods (EF: r=0.95, P<0.0001; PER: r=0.87, P<0.0001; TPER: r=0.84, P<0.0001; 1/3 FF: r=0.87, P<0.0001; PFR: r=0.92, P<0.0001; TPFR: r=0.89, P<0.0001). Bland-Altman plots did not reveal any significant degree of directional measurement bias in any of the comparisons of gated SPET data and RNV data. It is concluded that, in addition to the conventional LV systolic functional indices, our program accurately provides LV diastolic functional parameters from gated SPET. Also, this program will be useful for detecting LV diastolic dysfunction in various cardiac diseases before LV systolic dysfunction becomes evident.