Ventilation-perfusion scans were used in a prospective study to determine the prevalence of occult pulmonary embolus in proven deep venous thrombosis. Fifty-eight patients without symptoms of pulmonary embolism, but with venographically proven deep venous thrombosis, were subjected to chest radiographs, 99mTc macroaggregated-albumin perfusion scans, and 133Xe ventilation scans. Of the 49 patients with deep venous thrombosis proximal to the calf veins, 17 (35%) had high-probability scans. Of all 58 patients, only 12 (21%) had normal scans. When the study population was compared with a group of 430 patients described in reports of pulmonary perfusion in asymptomatic persons, a significantly higher percentage of high-probability scans was found in the study population with deep venous thrombosis. Baseline ventilation-perfusion lung scanning is valuable for patients with proven above-knee deep venous thrombosis.