Maximal stress thallium scans may prove to be 'normal' in some patients with triple-vessel disease due to global reduction in flow resulting in no focal perfusion defect. The aim of this study was to attempt to identify patients with global reduction in flow by estimating total thallium delivery to the left ventricle. Myocardial thallium uptake was calculated as a percentage of injected dose as a mean of three projections in 90 patients undergoing diagnostic arteriography and in 10 normal volunteers. These volunteers and nine patients who proved to have normal coronary arteries established a normal range. Values for myocardial thallium uptake were: 1.166 +/- 0.352% in normals (n = 19); 0.671 +/- 0.184% in patients with single- or double-vessel disease (n = 26); 0.708 +/- 0.245% in patients with triple-vessel disease (n = 55). Thallium scans were normal in 11 of 55 patients with triple-vessel disease and eight of 26 with single- or double-vessel disease. However, eight of these 11 and 5 of the eight proved to have abnormal myocardial thallium uptake. Thus combining the tests improved sensitivity from 76.5% for the scan alone to 94.5% for the scan and myocardial thallium uptake, with no reduction in specificity. Measurement of myocardial thallium uptake is a readily and reliably quantified parameter from a thallium scan which leads to enhanced detection of coronary artery disease.