DIAGNOSTIC-ACCURACY OF DOPPLER-ECHOCARDIOGRAPHY IN UNSELECTED PATIENTS WITH SUSPECTED PULMONARY-EMBOLISM

This study investigates the diagnostic value of echocardiography in pa tients with suspected pulmonary embolism. Doppler-echocardiography was performed in fifty consecutive patients, predominantly presenting in the emergency ward, with clinically suspected pulmonary embolism. Pati ents were classified as having or not pulmonary embolism by a sequenti al non-invasive strategy including lung scan, D-dimer measurement and lower limb venous compression ultrasonography, pulmonary angiography b eing performed in case of an inconclusive non-invasive work-up. The pr evalence of pulmonary embolism was 36% (18 of 50 patients). Right vent ricular dilatation on 2-D echocardiography associated to a tricuspid r egurgitation velocity greater than or equal to 2.7 m/s, corresponding to a pulmonary systolic pressure greater than or equal to 39 mmHg, wer e present in 12 of the 18 patients (67%) with and in two of the 32 pat ients (6.3%) without pulmonary embolism. They were, however, absent in five of the 18 patients (28%), in whom the definite diagnosis of pulm onary embolism was made. The combination of these both echocardiograph ic criteria yielded a sensitivity of 67% and a specificity of 94%, pos itive predictive value was 86% and negative predictive value was 83%. The diagnostic performance of these two combined echocardiographic cri teria, when present, permitted to reach in patients with a high clinic al pre-test probability of pulmonary embolism the post-test probabilit y values above 90%. On the other hand, the absence of these two Dopple r-echocardiographic criteria did not allow to exclude pulmonary emboli sm, except in presence of a low pre-test probability. The findings of our study show that Doppler-echocardiography in patients with high cli nical suspicion of pulmonary embolism may represent a potentially usef ul screening technique for the diagnosis of the disease permitting pro mpt initiation of treatment. However, the method does not allow to exc lude pulmonary embolism in all patients with intermediate or high clin ical suspicion of the disease. (C) 1998 Elsevier Science ireland Ltd.