The incidence of deep vein thrombosis following transfemoral temporary pacing was prospectively assessed in 20 consecutive patients using serial 125I-fibrinogen scanning, impedance plethysmography, and measurements of fibrin degradation products. Four patients underwent venography. Of the twenty patients, five (25%) had definite evidence of deep vein thrombosis, six demonstrated equivocal abnormalities, and nine had no evidence of thrombosis. Thrombosis was found only in the legs used for pacing. Of the five with deep vein thrombosis, one had positive findings in all the noninvasive tests. Venography confirmed the diagnosis in the other four; one had thrombosis of the femoral vein, one had thrombosis of several calf veins, and one had thrombosis of both the thigh and calf veins; the fourth patient had thrombosis of the entire deep venous system of the leg. This latter patient was the only one who had clinical evidence of venous thrombosis. The patients with deep vein thrombosis did not differ in clinical characteristics from those without thrombosis. There was no relationship between the occurrence of deep vein thrombosis and the number of venepunctures, the state of the electrode, the presence of the introducer sheath in the femoral vein, or the duration of pacing. In conclusion, deep vein thrombosis is a common complication of transfemoral temporary pacing and the high incidence of its occurrence should be considered before deciding upon the site of entry for a temporary pacemaker.