An assessment of the accuracy of hepatic and splenic size based upon a clinician's physical examination, a radiologist's impression and the actual liver and spleen volumes calculated by CT scanning

Hepatic height on coronal computed tomography images predicts total liver volume in European adults without liver disease

BACKGROUND

Hepatic volume measurement provides useful information in a large range of clinical situations.

AIMS

The goal of this study was to test the hypothesis that significant correlation exists between hepatic volume calculated using three-dimensional reconstruction of computed tomography (CT) data and hepatic height calculated using coronal reformation of CT images in European adult patients without liver disease.

METHODS

One hundred patients (50 men, 50 women; mean age, 47 years) without hepatic disease were included. Coronal and three-dimensional images of the liver were obtained using 64-section helical CT. Correlation between hepatic height and hepatic volume was searched for using the Pearson correlation test. Regression analysis was used to compare hepatic height and hepatic volume.

RESULTS

A strong and highly significant positive correlation was found between hepatic height and hepatic volume, with a correlation coefficient (r) of 0.767 (95%CI 0.672-0.837; r² = 0.588) (P < 0.001). The equation of the correlation line was y = 11.764x - 244 where y represents the hepatic volume in cm³ and x the hepatic height in mm.

CONCLUSION

Hepatic height as obtained on coronal CT image is a simple and rapid measurement that allows predicting hepatic volume in European adult patients without liver disease.

A novel canine model of portal vein stenosis plus thioacetamide administration-induced cirrhotic portal hypertension with hypersplenism

Current large animal models that could closely resemble the typical features of cirrhotic portal hypertension in human have not been well established. Thus, we aimed to develop and describe a reliable and reproducible canine cirrhosis model of portal hypertension. A total of 30 mongrel dogs were randomly divided into four groups: 1 (control; n = 5), 2 (portal vein stenosis [PVS]; n = 5], 3 (thioacetamide [TAA]; n = 5), and 4 (PVS plus TAA; n = 15). After 4-months modeling period, liver and spleen CT perfusion, abdominal CT scans, portal hemodynamics, gastroscopy, hepatic function, blood routine, the bone marrow, liver, and spleen histology were studied. The animals in group 2 (PVS) developed extrahepatic portosystemic collateral circulation, particularly esophageal varices, without hepatic cirrhosis and portal hypertension. Animals from group 3 (TAA) presented mild cirrhosis and portal hypertension without significant symptoms of esophageal varices and hypersplenism. In contrast, animals from group 4 (PVS + TAA) showed well-developed micronodular and macronodular cirrhosis, associated with significant portal hypertension and hypersplenism. The combination of PVS and TAA represents a novel, reliable, and reproducible canine cirrhosis model of portal hypertension, which is associated with the typical characteristics of portal hypertension, including hypersplenism.