Electron beam computed tomography and ventilation perfusion scintigraphy in the diagnosis of pulmonary embolism.

The purpose of this study, performed in patients with suspected pulmonary embolism (PE), was to compare V/Q scans and electron beam computed tomography (CT) scans on a patient-by-patient and segment-by-segment basis. Both a segment-based and a lobe-based analysis was performed in those patients positive for PE. The diagnosis of PE was assumed on the basis of a mismatch at V/Q scanning or a filling defect in a pulmonary vessel at contrast-enhanced electron beam CT. In 37/45 patients (24 female, 21 male, 58+/-16 years) with suspected PE, the diagnosis of PE was confirmed or excluded by both modalities, resulting in a correlation of 82% between electron beam CT and V/Q scanning. In the 28 patients positive for PE according to one or both modalities, 504 segments were evaluated. Of these 504 segments, 248 (nearly 50%) showed perfusion defects on V/Q scans, of which only 90 (36%) displayed emboli at electron beam CT. Overall, a total of only 135 of the 504 segments (27%) were abnormal at electron beam CT. More than 50% of the patients with discrepant results did not show an embolus at electron beam CT. It is concluded that there is a good correlation (82%) between V/Q scanning and electron beam CT on a patient-by-patient basis but a markedly less good correlation (62%) in a segment-based analysis.

Electron beam tomography of interpulmonary saddle embolism: extent and vascular distribution.

PURPOSE: The purpose of this work was to assess morphology and secondary signs of interpulmonary emboli extending across the bifurcation of the main pulmonary artery (PA). METHOD: Retrospective evaluation of 780 electron beam tomographic studies of the chest yielded 17 cases of interpulmonary saddle embolism. Length, diameter, vascular distribution of the emboli, and secondary findings were prospectively assessed by two blinded reviewers. Follow-up studies were carried out in 12 of 17 patients (71%). RESULT: Mean total length of the emboli was 46.6 cm (range 20.9-81.5 cm). The mean diameter of the saddle embolus was significantly smaller at the level of the bifurcation than in the left or right PA (4.5, 7.7, and 7.4 mm, respectively; p < 0.01). Dilatation of the right heart was found in 10 of 17 cases (59%). At follow-up, the saddle state was no longer present in 8 of 12 patients (67%). CONCLUSION: Interpulmonary saddle emboli appear to be a transient form of acute pulmonary embolism, the site of predilection for rupture of the embolus being the level of the bifurcation. Their frequency may therefore be underestimated.