Diagnostic value of Doppler assessment of the hepatic and portal vessels and ultrasound of the spleen in liver disease

OBJECTIVESTo investigate the clinical utility and the intra-observer and inter-observer variability of Doppler ultrasound assessment of the hepatic and portal vessels along with measurement of spleen size in the diagnosis of chronic liver disease and cirrhosis. METHODS AND MATERIALSUltrasound measurements of portal vein diameter (PVD), portal vein velocity (PVV), hepatic arterial resistance index (HARI), hepatic vein profile (HVP), and spleen size were obtained in 49 controls and 45 patients with liver disease (23 with primary biliary cirrhosis, 22 with hepatitis C) by two experienced observers, who each performed three blinded measurements of each variable. Control values were derived from normal hospital workers. Percutaneous liver biopsies in 41 of the patients showed cirrhosis (14 patients), moderate/severe fibrosis (13 patients), and early disease (14 patients). RESULTSSeventy-one percent of cirrhotic patients had splenomegaly (> 13.6 cm). The spleen size was significantly larger in cirrhotics (16.0 cm) than in non-cirrhotics (13.0 cm, P < 0.009) and healthy controls (10.7 cm, P < 0.00005), and was the only independent predictor of cirrhosis, with a threshold of 15 cm predicting cirrhosis with a specificity of 98%, positive predictive value of 93%, sensitivity of 57% and negative predictive value of 80%. HVP was abnormal in 76.9% of cirrhotics, 57.7% of non-cirrhotics and 2.1% of controls (P < 0.04). However, the mean PVV, PVD and HARI were no different between controls and patients or between cirrhotic and non-cirrhotic liver disease. There was significant inter-observer variability for PVV, but intra-observer and inter-observer variability was acceptable for the other measurements. CONCLUSIONSSplenomegaly size and abnormal HVP are useful predictors of chronic liver disease and cirrhosis, and both can be measured reliably and reproducibly. However, Doppler measurements of PVV, PVD and HARI are not useful in distinguishing patients with chronic liver disease from normal controls.