[Pulmonary embolism in leg and pelvic vein thrombosis. Results of a prospective study - diagnosis, frequency, nuclearmedical and clinical appearance (author's transl)].

In a prospective study 169 patients with clinically suspected leg/pelvic-vein thrombosis were tested by 131 I fibrinogen uptake-test and radionuclide venography for confirmation of this diagnosis and also by a nuclearmedical lung investigation. In all cases of an abnormal perfusion scan a supplementary ventilation scan was performed. The perfusion scan has a sensitivity for pulmonary embolism of near 1.0 but its specificity is only 0.57. An additional ventilation study (133 Xenon, 81m Krypton) improves the specificity to 0.95. A mismatch of regional ventilation and perfusion is the nuclearmedical substrate of pulmonary embolism. The diagnosis of thrombosis was confirmed in 105 of 169 cases (62%). Thrombosis was located in the lower legs in 56%, in the thigh veins in 23% and in the external iliac veins in 21%. From 105 patients with leg/pelvic-vein thrombosis 60 (57.1%) had pulmonary emboli, from 64 patients with negative tests concerning thrombosis only 3 (4.7%) (p < 0.001). Pulmonary emboli were present in 46% when thrombosis was located in the lower legs, in 67% when thigh veins and in 77% when pelvic veins were involved. The average frequency of 57% emboli found in patients with leg and pelvic vein thrombosis agrees with data from pathologic-anatomical studies. An analysis of these patients with embolism showed that 70% of them were over 70 years old, that 52% of the emboli originated from thigh and pelvis and 43% from the lower leg and that 59% had no clinical signs of embolism. 80.4% of the patients had multiple perfusion defects (up to 9) which correlated in size with the severity of the clinical symptoms and which were about equally distributed in both lungs. Larger perfusion defects occure more frequently with thromboses of the thigh and pelvis than in thromboses of the lower leg. According to the chest x-ray pulmonary emboli were suspected to only 6 of 26 patients with clinical evidence of embolism (23%).