Lung images in which perfusion defects correspond to chest radiograph (CXR) opacities in size and location are generally classified as having an intermediate probability for pulmonary embolism, while those studies in which the perfusion defects are considerably smaller than the CXR opacities are classified as having a low probability for pulmonary embolism. A case is presented in which, by imaging a patient in both the erect and supine positions, the authors were able to change the interpretation of the study from intermediate probability (perfusion defect = CXR opacity) to low probability (perfusion defect less than CXR opacity).