Pulmonary embolism is commonly fatal, yet notoriously difficult to detect. Diagnosis often relies on the ventilation-perfusion radionuclide scan, which itself is frequently equivocal. It has been suggested that if the equivocal ventilation-perfusion scan is interpreted in the light of clinical information, diagnostic accuracy can be improved. However, which features in the history should be considered? In this study of 197 patients undergoing ventilation-perfusion scanning, the clinical data of the 98 patients with either high-probability or normal scans were compared to the scan findings. The presence of a deep vein thrombosis was significantly associated with a high probability scan, whereas the presence of constant chest pain was significantly associated with a negative scan. Classical symptoms for pulmonary embolism, namely pleuritic chest pain and hemoptysis, were poor predictors of high-probability scans. Consequently, the authors advise considerable caution when using the clinical data to aid the interpretation of the equivocal lung scan in the individual case.