Giannini, Mari [UNESP]Rollo, H. A. [UNESP]Maffei, F. H.A. [UNESP]Carvalho, L. R. [UNESP]2022-04-282022-04-282008-01-01International Angiology, v. 27, n. 6, p. 466-474, 2008.0392-9590http://hdl.handle.net/11449/225418Aim. This study aimed at assessing the accuracy of ultrasound (US) in the diagnosis of recent deep vein thrombosis (DVT) in an experimental study in dogs. Methods. Design: blinded and randomized experimental study. Twenty dogs were randomly divided in two groups: control group (CG) and thrombosis group (TG). US was performed in the pre- and postoperative period. Phlebography was performed immediately prior to the postoperative US. After the second US, a surgery was performed to detect whether thrombus was present or not. US results were compared to those of phlebography and surgical findings. Results. In all dogs, inferior vena cava (IVC) was compressible. The relations of IVC diameter with the aorta were higher (P<0.005) in TG than in CG. Spectral Doppler in spontaneous breathing, tissue harmonic imaging, power Doppler and B flow showed sensitivity, specificity and accuracy of 1. Phlebography presented sensitivity of 90%, specificity of 80% and accuracy of 85%, when compared to surgical finding. Conclusion. For the diagnosis of recent DVT in the experimental model used, venous compressibility proved to be inefficient. The ratio of IVC diameter to aorta, when increased, suggests thrombosis. The use of new US technological advances increases accuracy. Phlebography was less accurate than US.466-474engUltrasonographyVenous thrombosis, diagnosisVenous thrombosis, therapyArtigo2-s2.0-60549116303