To evaluate the utility of the alveolar-arterial (A-a) oxygen gradient in the diagnosis of acute pulmonary embolism (PE), a retrospective analysis was done of consecutive emergency department patients who underwent pulmonary angiography for the presumed diagnosis of acute PE. Patients were categorized into two groups depending on the presence or absence of prior cardiopulmonary disease. Arterial blood gas samples were chosen for analysis only if obtained when the PE first was suspected clinically (before lung scans or angiograms) and the patient was breathing room air. A total of 152 patients met all study criteria; 59 patients (39%) had angiographically documented emboli. In comparison with the study patients in whom PE was excluded, there was no significant difference in mean PaO2 (64 v 67 mm Hg) or A-a gradient (39 v 36 mm Hg). Various combinations of the A-a gradient and blood gas levels failed to exclude PE in more than 35% of patients with no prior cardiovascular disease and in 25% of patients with prior cardiovascular disease. The A-a gradient did show a linear correlation with the severity of the PE, as assessed by the PaO2 (r = -0.87) and pulmonary artery mean pressure (r = 0.63). These results indicate that the A-a oxygen gradient, in combination with blood gas levels, may contribute to the formulation of a clinical assessment. However, these laboratory parameters are of insufficient discriminant value to permit exclusion of the diagnosis of PE.