Correlation of duplex sonography findings and portal pressure in 375 patients with portal hypertension.

Abstract

The purpose of this study was to determine the potential usefulness of duplex sonography in the grading of portal hypertension. Duplex sonography of the portal vein system and measurement of the portal pressure and portosystemic pressure gradient were performed in 375 patients before placement of transjugular intrahepatic portosystemic shunts. Subgroups included patients with recent variceal bleeding (n = 296) and patients with refractory ascites without previous variceal bleeding (n = 79). A matched cohort of 100 patients without portal hypertension was also examined. Differences between the groups in portal and splenic vein diameter, flow velocity, congestion index, and hepatic arterial resistive index were assessed using the Wilcoxon rank sum test. Compared with healthy individuals, our patients had an increased portal vein diameter (+30%, p < .001), decreased portal vein flow velocity (-44%, p < .001), and increased congestion index (+185%, p < .001). A portal vein diameter greater than 1.25 cm or a portal vein flow velocity less than 21 cm/sec indicated portal hypertension with a sensitivity and specificity of 80%. If the congestion index exceeded 0.1, portal hypertension was diagnosed with a 95% sensitivity and specificity. The portal pressure and gradient correlated only weakly (r < .2, p < .05) with sonographic variables. Using multivariate analysis, subgroups with variceal bleeding or refractory ascites did not show differences in hemodynamics, including pressures. Duplex sonography contributes to the diagnosis of portal hypertension but does not allow its grading. Similarity of portal hemodynamics between patients with variceal bleeding and patients with refractory ascites suggests that additional factors determine the respective clinical presentation.