Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis

Objective. —To determine the frequency of pulmonary embolism in patients admitted for treatment of deep venous thrombosis. Design. —An open, multicenter, dose-ranging study to assess the safety and pharmacokinetic characteristics of tissue-type plasminogen activator in deep venous thrombosis and pulmonary embolism. Perfusion and ventilation lung scans, chest roentgenograms, and venograms (in deep venous thrombosis) or pulmonary angiograms (in pulmonary embolism) were obtained before and 24 hours after inception of therapy. Heparin therapy was then administered. Settings. —Five tertiary-care hospitals. Patients. —All patients with suspected deep venous thrombosis or pulmonary embolism seen from August 1987 through November 1988 entered the study if they met inclusion criteria and if the diagnosis was confirmed by venogram (deep venous thrombosis) or pulmonary angiogram (pulmonary embolism). Interventions. —All patients received tissue-type plasminogen activator followed by intravenous heparin therapy. Main Outcome Measures. —The primary measure was the frequency of pulmonary embolism in patients with deep venous thrombosis who had no symptoms of pulmonary embolism. This was not the original purpose of the study but emerged as an important finding as the data were analyzed. Results. —Nearly 40% of patients with deep venous thrombosis who had no symptoms of pulmonary embolism had evidence of pulmonary embolism based on ventilation-perfusion scan and chest roentgenogram findings. Conclusions. —Because all of those considered to have embolism had so-called high-probability scan results, the frequency of embolism reported likely represents the minimum incidence of pulmonary embolism in patients with deep venous thrombosis who have no embolic symptoms. These data emphasize that venous thromboembolism is one disorder. (JAMA. 1994;271:223-225)