Objectives: Duplex ultrasonography performance in detecting embolic foci has not been proven satisfactory compared with phlebography or autopsic findings. In case of suspected pulmonary embolism, the embolic focus is only discovered in 11 to 18% of the cases compared with more than 30% with phlebography. For overt acute pulmonary embolism, the discovery rate is in the 30 to 45% range versus 70 to 80% with phlebography or autopsy findings. This discrepancy might result from the fact that duplex ultrasonographic explorations are generally limited to the deep collectors at the cruropopliteal level. The purpose of this study was to assess the prevalence of duplex ultrasonography detected venous thrombosis in patients with suspected or acute pulmonary embolism when the exploration includes the entire venous system from the inferior vena cava to the ankles and examines not only the deep collectors but also the muscle and superficial networks.
Material and methods: This study included all patients with suspected pulmonary embolism referred to the emergency unit from January 1, 1995 through December 31, 1998. The patients' hospital files were used to determine the suspected pulmonary embolism population. The acute pulmonary embolism population was defined as the patients whose files contained documented proof of pulmonary embolism (highly probable ventilation/perfusion pulmonary scintigraphy, positive pulmonary angiography, positive proximal angioscan). Thrombosis of the deep venous collectors with or without associated superficial or muscular localization was classed as "deep venous thrombi" and superficial or muscular thrombosis without involvement of the deep collectors was classed as "other venous thrombi". Subpopliteal thrombosis was classed as distal and popliteal or suprapopliteal thrombosis as proximal.
Results: The suspected pulmonary embolism group included 352 patients, 118 men and 234 women aged 67.6 +/- 15.4 and 70.8 +/- 20.0 years respectively (m +/- SD). The acute pulmonary embolism group included 60 patients, 17 men and 43 women aged 66.2 +/- 12.5 and 69.7 +/- 16.6 years respectively. Overall prevalence of duplex-ultrasound detected venous thrombosis was 30.4% (107/352) (95%CI: 25.6-35.2) in the suspected pulmonary embolism group and 80% (48/60) (95%CI: 69.9-90.1) in the acute pulmonary embolism group. Deep venous thrombi reaching the collectors and proximal thrombi predominated. Prevalence of "other venous thrombi" and distal venous thrombi were 6.5% (23/352) and 11.4% (40/352) respectively in the suspected pulmonary embolism group and 15.0% (9/60) and 26.7% (16/60) in the acute pulmonary embolism group. The frequency of asymptomatic venous thrombosis of the lower limbs, irrespective of the localization, was 42.1% (45/107) in the suspected pulmonary embolism group and 52.1% (25/48) in the acute pulmonary embolism group.
Conclusions: The prevalence of duplex-ultrasonography detected venous thrombosis in patients with suspected or proven pulmonary embolism found in this series was equivalent to the rates reported in phlebography and autopsy series. The prevalence was higher than usually reported for duplex-ultrasonography studies limited to the cruro-popliteal level. The difference came from the "other venous thrombi" and "distal deep venous thrombi" discovered by exploring the superficial and muscular networks and the calves. This study demonstrates the contribution of duplex-ultrasonography to the diagnostic strategy for pulmonary embolism.