Prolonged impairment of left ventricular (LV) systolic function following exercise induced ischaemia has been well demonstrated. The objective of this study was to examine the effect of exercise induced ischaemia on the post-stress LV diastolic function in patients with coronary artery disease (CAD). Seventy-four subjects with known or suspected CAD underwent gated myocardial single photon emission computed tomography (SPECT) 1 h after administration of 99mTc tetrofosmin according to a standard same day exercise rest protocol. LV volumes and ejection fractions (LVEFs) were determined by the Cedars-Sinai program. Fourier transformation of the gated SPECT volume curve was performed retaining the fourth order harmonics, and peak filling rate (PFR) and time-to-PFR (TPFR) were calculated from the derivative curve. In patients with exercise induced ischaemia (n =26), 1 h post-stress PFR (2.66+/-0.75 s(-1)) and TPFR (119+/-12 ms) were significantly impaired in comparison to the resting PFR (3.06+/-0.74 s; P=0.0002) and TPFR (114+/-10 ms; P=0.03), respectively. In normal subjects (n =26) and in patients with infarction (n =22), the post-stress indices were similar to the resting values. When reduction of PFR or LVEF greater than the variability (2SD) of differences between the post-stress and resting values in the normal group was defined as significant impairment, six of the 26 ischaemic patients (23%) had such changes in PFR. All these patients exhibited severe ischaemia and five of them had simultaneous systolic impairment. Only one (4%) of the normal subjects and none of the patients in the infarction group showed such impairments. Stepwise logistic regression analysis of stress, scan and coronary variables revealed that the summed reversibility score, a scintigraphic index of ischaemic severity, was the only determinant of post-stress changes in LVEF and PFR. In conclusion, exercise induced LV diastolic impairment persists for a prolonged period after resolution of the ischaemic episode. The incidence and magnitude of the diastolic impairment are determined by the severity of the exercise provoked ischaemia.