To determine the value and limitations of quantitative analysis of thallium-201 imaging after intravenous dipyridamole in combination with low level exercise, 81 patients with suspected coronary artery disease (CAD) were evaluated prospectively. The results of quantitative analysis were compared with the results of visual analysis. All patients underwent coronary arteriography and left ventricular angiography. Significant CAD was present in 59 patients (73%); multivessel CAD was observed in 33 patients (42%). Mild side-effects such as headache, vertigo and nausea were experienced by 12 patients (15%). To establish 'test-specific' normal limits in quantitative analysis of uptake and washout of thallium-201 after dipyridamole infusion with low level exercise we studied 20 healthy volunteers with a likelihood of CAD less than or equal to 1%. Sensitivity and specificity of quantitative analysis for overall detection of significant CAD were 76% and 73%, respectively vs 78% and 86%, respectively with visual analysis (P = NS). Sensitivity for the detection of multivessel CAD was slightly higher with quantitative analysis (73%) than with visual analysis (70%), but the specificity was lower (75% vs 92%) (P = NS). Sensitivity of quantitative analysis in relation to vessels involved was slightly lower than with visual analysis (RCA and LAD: P = NS; LCX: P less than 0.05). Of 14 patients with false-negative results, four had multivessel CAD. Thus, quantitative analysis after the combination of dipyridamole infusion with low level exercise did not improve the diagnostic value as assessed by semi-quantitative visual analysis.