The effect of left ventricular (LV) systolic function on left atrial (LA) filling was studied using radionuclide angiography (RNA) and echocardiography in 21 normal subjects and in 40 patients with various heart desease who had no regurgitant or shunt flow. The LV and LA time-activity curves(TAC) and these first-derivative curves werederived from RNA. The peak rate of LA filling during the atrial filling phase was normalized to the LA maximum counts and was defined as the LA peak filling rate (LAf-PFR). In all patients and normal subjects, the LV peak ejection rate (LVs-PER) occurred at a time close to that of Laf-PFR ((132±22msec (mean ±SD) vs 139±34msec, NS)). With use of the following equations, the peak rates of the LV emptying or LA filling determined by the RNA were converted from counts/sec to ml/sec by equating the echocardiographically-derived LV end-diastolic volume or LA maximum volume to the LV end-diastolic counts or the LA maximum counts, respectively. LVs-max dV/dt (ml/sec) =(LVs-PER)×(LV end-diastolic volume) LAf-max dV/dt (ml/sec)=(LAf-PFR)×(LA maximum volume.) There was a significant positive correlation between the LVs-max dV/dt, an index of LV systolic function, and the LAf-max dV/dt, an index of LA filling function (r=0.58, p<0.001) in all patientws and normal subjects. This indicates that the LA peak filling rate increase ptogressively as the LV peak ejection rate. Thus, these results suggest that the LA filling might be affected by the LV systolic function.