ABSTRACT
We have study prospectively cases of non severe pulmonary embolism in a pulmonary department by an analysis based strategy associating radio-clivical probability, venous ultra sonography, D-Dimers value followed, if no performed diagnosis, by pulmonary scintigraphy or angio-CT scan. 64 cases of pulmonary embolism suspicion have been hospitalised in our department between October 1998 and July 2001; 40 patients was included in our study and have been classified in 3 groups regarding pre test clinical probability. Anticoagulant treatment has been initialised only in the third group [probability > 80%]Clinical probability associated with venous ultra sonography and D- Dimeres value allow or exclude pulmonary embolism diagnosis in 27 patients. In the others, scintigraphy and angio CT scan were necessary for establishing diagnosis. Application of this algorithm allow diagnosis of pulmonary embolism in 29 patients and exclude this pathology in the other 11. None of this patient complained from recurrent thrombo embolic accident during 17 to 42 months observance period