This study investigated the effect of vasopressin on portal-systemic collaterals in portal hypertensive rats and the influence of nitric oxide (NO) and prostaglandin on the responsiveness of collateral vessels to vasopressin. The vascular responsiveness to graded concentrations of vasopressin was tested with or without the incubation of n -nitro-L-arginine (NNA) (100 mol/L) and/or indomethacin (10 mol/L) in perfused collateral vascular beds of rats with portal hypertension induced by partial portal vein ligation. In addition, concentration-response curves to vasopressin with incubation of a vasopressin V 1 receptor antagonist d(CH 2 ) 5 Tyr(Me) arginine vasopressin and concentrationresponse curves to a V 2 receptor agonist 1-desamino-8-Darginine vasopressin were performed. Vasopressin significantly increased the perfusion pressure of collaterals, and this effect was suppressed by the addition of the V 1 receptor antagonist. Perfusion with the V 2 receptor agonist had no effect on the collaterals. Incubation with NNA, indomethacin, or both significantly potentiated the response of collaterals to vasopressin. In addition, the pressor response to vasopressin in the combination group was significantly higher than that in the NNA-alone group. The results show that vasopressin produces a direct vasoconstrictive effect on the portal-systemic collaterals of portal hypertensive rats. This effect is mediated by the vasopressin V 1, but not V 2 , receptors. The attenuation of the response to vasopressin by NO and prostaglandin suggest a function role of both mediators in the regulation of the portal-systemic collateral circulation in portal hypertensive rats. (HEPATOLOGY 1999;30:630-635.)Vasopressin, an agent that decreases portal and collateral blood flow and portal pressure, 1 has been used in cirrhotic patients for the management of esophageal variceal bleeding. 2 It is a powerful vasoconstrictor and exerts varying degree of contractile effect over arteries and veins. 3 Perfusion studies of splanchnic vascular beds have shown that vasopressin markedly increases the vascular resistance of mesenteric arteries but does not alter the venous resistance. 4 Therefore, the constriction of splanchnic arteries accompanied by the reduction of splanchnic and portal-systemic collateral blood flow have been postulated to be the major mechanisms by which vasopressin alleviates portal hypertension and controls variceal hemorrhage in patients with cirrhosis. However, it is uncertain if the effect of vasopressin on reducing collateral blood flow is also mediated by a direct vasoconstrictive effect on the collateral vascular beds.Nitric oxide (NO) and prostacyclin (PGI 2 ) are endogenous vasodilators synthesized by the vascular endothelium. 5,6 Both of them modulate mesenteric vascular tone and are important contributors to splanchnic hyperemia in portal hypertensive rats. [7][8][9][10][11] In addition, NO plays a role in modulating collateral vascular resistance, 12 and increased PGI 2 activities have been observed in systemic circulatio...

The present study investigated plasma levels of interleukin-8 (IL-8) in patients with post-hepatitic cirrhosis and correlated it with the severity of liver diseases and haemodynamic parameters. Plasma IL-8 levels were significantly higher in 57 post-hepatitic cirrhotic patients (7.5 +/- 1.8 pg/mL; P < 0.005) than those in 41 healthy subjects (2.0 +/- 0.2 pg/mL). Elevated (> 5 pg/mL) plasma IL-8 levels were found in up to 30% of cirrhotic patients. In cirrhotic patients, plasma IL-8 levels progressively increased in relation to the severity of liver dysfunction (4.5 +/- 1.0, 4.9 +/- 1.4 and 20.5 +/- 8.3 pg/mL for Pugh's class A, B and C, respectively; P < 0.005). A significant correlation was observed between plasma IL-8 levels and serum bilirubin levels (r = 0.72; P < 0.001). There were no differences in the hepatic venous pressure gradient (15.4 +/- 1.1 vs 15.1 +/- 0.9 mmHg; P > 0.05) and systemic vascular resistance (1119 +/- 118 vs 1199 +/- 54 dyn.s/cm5; P > 0.05) between cirrhotic patients with and without elevated plasma IL-8 levels. In addition, plasma IL-8 levels did not correlate with hepatic venous pressure gradient (r = 0.26; P > 0.05) and systemic vascular resistance (r = -0.24; P > 0.05). These results demonstrate that plasma IL-8 levels are increased in patients with post-hepatitic cirrhosis. The severity of liver cirrhosis is an important factor for the occurrence of enhanced IL-8 levels. IL-8 does not play a role in the hyperdynamic circulation observed in patients with post-hepatitic cirrhosis.

Focal hepatic fatty infiltration may be mistaken for hepatic neoplasm on ultrasonography. A sonographic feature mimicking "space occupying lesions" was identified in 41 cases with focal fatty liver, which led to the recognition of four sonographic patterns: (1) Ten patients in whom the livers were characterized by irregular configurations of hyperechoic and hypoechoic areas. Although confusing to inexperienced sonographers, all of them were properly interpreted by experienced sonographers. (2) One patient in whom the hepatic sonography revealed a hyperechoic nodule, but which was correctly diagnosed by computed tomography (CT) scan. (3) Six patients in whom the livers were characterized by multiple confluent hyperechoic lesions, which were all misinterpreted by sonographers and misinterpreted following CT scans in 4 cases and angiography in 3 cases. (4) Twenty-four patients in whom the livers revealed focal spared areas, which was misdiagnosed by sonographers in 14 cases, but correctly diagnosed by CT scan in all 24 cases.

ABSTRACT— A prospective survey of primary hepatocellular carcinoma (PHC) was conducted in a rural area of Taiwan using a two‐site enzyme immunoassay for alphafetoprotein (AFP) in dried blood samples collected on filter paper. Of 1894 men over 40 years of age who were tested, 20 (1%) had AFP levels of greater than 20 ng/ml of blood on screening. Nineteen of these men received ultrasound examinations, and small PHCs were detected in 4 (21%). The remaining 15 cases had other types of hepatobiliary tract diseases, and 6 (40%) were also serum HBsAg positive. They should all be closely followed up by AFP determination and ultrasound examination of the liver. In contrast to the low resection rate of PHC in symptomatic patients who were admitted through the outpatient clinic during the period of this survey, the four cases with small PHC discovered by AFP screening had their tumors successfully removed (5/57 vs. 4/4, p<0.005). Our preliminary results showed that this method is a simple, sensitive and convenient assay for AFP and may be used as a first‐line screening test in mass population surveillance programs for PHC, particularly in areas where PHC is highly prevalent.

The association between prolonged bleeding time and hepatocellular carcinoma (HCC) has not been well studied. We investigated whether bleeding time is prolonged in cirrhotic patients with HCC and studied the role of clinical characteristics, tumour size, and laboratory data in predicting bleeding time prolongation. After excluding patients that presented with blood dyscrasia and uraemia, 58 cirrhotic patients with HCC, 106 cirrhotic patients without HCC, and 44 age-and sex-matched healthy subjects were included in the study. Bleeding time, imaging studies, clinical characteristics and biochemical data were obtained for every patient. Cirrhotic patients with and without HCC had longer bleeding times (554 +/- 32 s, respectively) compared with healthy controls (357 +/- 13 s, P < 0.05). Hepatocellular carcinoma patients with a large tumour burden (> 5 cm in diameter) had a significantly longer bleeding time than those patients without (663 +/- 105 vs 376 +/- 23 s, respectively, P < 0.05). After excluding patients with a platelet count < or = 80 000/mm3, cirrhotic patients classified as Child-Pugh's grading A and with a large tumour burden had longer bleeding times(580 +/- 87 s) than patients with a small tumour burden (< or = 5cm in diameter) and cirrhotic patients without HCC (371 +/- 22 and 416 +/- 29 s, respectively, P < 0.05). In cirrhotic patients with HCC, higher serum bilirubin levels, a Child-Pugh's grading C, and a tumour size > 5 cm in diameter were found to be significant predictors for prolonged bleeding time on univariate analysis. On multivariate analysis, both tumour size > 5 cm in diameter and a Child-Pugh's grading C (odd's ratio, 95% confidence interval and P value were measured as 38.5, 2.8-534.7, < 0.001, and 10.5, 0.9-117.6, 0.02, respectively) were the significant independent predictors. A significant correlation existed between tumour diameter and bleeding time (r = 0.44, P < 0.01). In conclusion, these results suggest that prolonged bleeding time may be categorized as a new clinical manifestation in patients with HCC. In addition to cirrhosis, HCC itself may also participate in the pathogenesis of bleeding time prolongation.