RESUMEN
The reference standard for portal venous pressure measurement which is clinically important for estimating the feasibility of resection of hepatocellular carcinoma is the hepatic venous pressure gradient, which is invasive and expensive. The present study evaluated the noninvasive parameters for assessment of portal hypertension in Child A patients with hepatocellular carcinoma on top of hepatitis C virus. A total of 112 patients were subjected to clinical assessment, biochemical assay, ultra- sonographic Doppler study, triphasic spiral abdominal computed tomography, upper gastrointestinal endoscopy and hepatic venous pressure gradient measurement. According to hepatic venous pressure gradient measurement, they were classified into groups: GI: 58 patients with hepatic venous pressure gradient <10 mmHg and GII: 54 patients with hepatic venous pressure gradient >/= 10 mmHg. Significant variables in univariate analysis were included in a multivariate analysis to establish a model for prediction of clinically significant portal hypertension. Results showed that portal vein diameter >/= 1.3 cm, mono or biphasic pattern of flow in hepatic veins and Giannini index /= 10 mmHg. A model with highest likelihood ratio and good fitness was created. This prediction model was displayed by the receiver operating characteristic curve and under the curve area was 0.969 [0.938-1]