To determine whether exercise-induced ischemia impairs left ventricular diastolic filling in the postischemic period in humans, 101 men (mean age 57 +/- 10 years) were studied before and 2 h after a symptom-limited thallium-201 tomographic treadmill with pulsed Doppler echocardiography of mitral valve inflow. In the postischemic period 2 h after exercise, diastolic filling was significantly impaired in the ischemia group (reversible thallium defect; n = 24) as reflected by a decrease in the peak early filling velocity (44.5 +/- 10.1 to 39.9 +/- 9.9 cm/s, p < 0.01), peak early to atrial filling velocity ratio (0.91 +/-0.27 to 0.76 +/- 0.25, p < 0.001), and deceleration rate of early filling (281 +/- 104 to 245 +/- 86 cm/s2, p < 0.01). Similar alterations in the postischemic period occurred in the myocardial infarction-ischemia group (partially reversible defect; n = 28) as seen by a decrease in the peak early filling velocity (47.6 +/- 11.6 to 41.8 +/- 12.0 cm/s, p < 0.001), peak early to atrial filling velocity ratio (0.84 +/- 0.21 to 0.68 +/- 0.18, p < 0.001), and early time-velocity integral (7.06 +/- 1.78 to 5.64 +/- 2.07 cm, p < 0.001). In the control group (no defects; n = 33) and myocardial infarction group (fixed defect; n = 16), diastolic filling was unchanged in the postexercise period. Heart rate and blood pressure were unchanged post-exercise in all groups. Exercise-induced ischemia impairs diastolic filling in the postischemic period in humans. This most likely reflects impairment of diastolic function, but alterations of loading conditions cannot be completely excluded. It is possible that recurrent episodes may chronically depress diastolic function and contribute to morbidity in subjects with coronary artery disease. The mechanism of how exercise-induced ischemia "stuns" or impairs diastolic filling remains unclear but requires further study.