To determine the relationships between perfusion scan defect and angiographic severity (Miller index) in acute pulmonary embolism, we analysed examinations obtained before and after thrombolytic therapy in 34 consecutive patients free from underlying cardiopulmonary disease. The overall agreement between the two techniques was excellent (r = 0.82; mean absolute difference = 2.8%), although when embolic involvement was extensive (greater than 50% angiographic obstruction), the perfusion scan moderately underestimated (4%) the defect seen angiographically. These findings suggest that the pulmonary lung scan is a reliable method of assessing the initial pulmonary vascular obstruction as well as of quantifying any changes induced by or associated with the treatment.