Sixty-eight patients clinically suspected of having (33), or at high risk for (35), deep venous thrombosis were studied with contrast venography, radiofibrinogen leg scanning, and impedance plethysmography as well as ventilation and perfusion lung scans. Thrombosis limited to the veins in the calf of the leg (unilateral or bilateral) was shown by venography in 12 patients. None of these patients had clinical symptoms or scan results indicating embolism. Fifteen patients had thrombosis involving proximal (thigh) as well as distal (calf) veins by venography. Eight had scan evidence of embolism, although only one was symptomatic. The combination of radiofibrinogen and impedance tests allows accurate detection of both the presence and location of deep venous thrombosis. The availability of sensitive and specific, noninvasive methods for detecting and localizing venous thrombosis, as well as the apparently low embolic risk of calf-only thrombosis may condition future approaches to prophylaxis and treatment of patients with or at high risk for deep venous thrombosis.