Search

Background: Patients with unresectable hepatoma and acute esophageal variceal bleeding have extremely high rates of recurrent bleeding and mortality. This controlled study evaluates the feasibility and potential benefit of maintenance endoscopic variceal ligation in these patients. Methods: Patients with unresectable hepatoma and acute esophageal variceal bleeding underwent emergent endoscopic variceal ligation. After hemostasis, patients were randomized to undergo maintenance or esophageal variceal ligation (EVL) as necessary (demand ligation). Results: Fifty-four patients underwent maintenance EVL and 55 demanded EVL. One or more subsequent EVL session could be performed in only 30 patients (55.6%) in the maintenance group (actual maintenance ligation). Logistic regression analysis found that hepatic function determines the feasibility of maintenance ligation (Child-Pugh's A+B vs. C, OR 23.00: 95% CI [5.26, 100.66]). The survival and recurrent bleeding rates were similar in both groups. A subgroup analysis of patients with Child-Pugh's A and B hepatic reserve in both the maintenance EVL group (n = 24) and demand EVL group (n = 25) was performed to assess the potential benefit of maintenance ligation. Maintenance ligation reduced the rate of recurrent bleeding compared with demand ligation (p = 0.043). Cox regression showed that portal vein thrombosis and tumors in both hepatic lobes were also factors together with EVL that determined recurrence of bleeding. Survival was similar in both groups. Conclusions: Maintenance ligation is feasible in patients with unresectable hepatoma and variceal hemorrhage if they have a good hepatic reserve. Maintenance ligation might lower the rate of recurrent bleeding in this subgroup of patients. (Gastrointest Endosc 2001;54:18-23.)

Background: Liver cirrhosis with portal hypertension is associated with hyperdynamic circulation characterized by generalized vasodilatation and increased cardiac output and regional blood flows. Patients with liver cirrhosis present with increased levels of interleukin-6 (IL-6), which may inhibit vascular smooth-muscle contraction. We investigated whether increased plasma IL-6 levels contribute to the pathogenesis of hyperdynamic circulation observed in cirrhotic patients and whether they are correlated with plasma tumor necrosis factor-α (TNF-α) and endotoxin concentrations. Methods: In 58 consecutive cirrhotic patients and 34 healthy subjects the plasma concentrations of TNF-α and IL-6 were measured with enzyme-linked immunosorbent assay, and endotoxin determinations with a limulus assay. In addition, 52 cirrhotic patients underwent a hemodynamic study using Swan-Ganz catheterization. Results: Plasma TNF-α, IL-6, and endotoxin levels were significantly higher in cirrhotic patients than in healthy subjects (7.3 ± 0.2 versus 5.8 ± 0.1 pg/ml, 6.4 ± 0.8 versus 2.0 ± 0.2 pg/ml, and 7.6 ± 1.2 versus 2.8 ± 0.3 pg/ml, respectively; p < 0.01). In cirrhotic patients the plasma levels of TNF-α, IL-6, and endotoxin progressively increased in relation to the severity of liver dysfunction (graded by Pugh's classification). A significant correlation was observed between plasma TNF-α and IL-6 levels (r = 0.48, p < 0.001), whereas no correlation was observed between plasma endotoxin levels and plasma TNF-α and IL-6 levels. Plasma IL-6 levels correlated negatively with systemic vascular resistance in patients with cirrhosis (r = 0.5, p < 0.01). Conclusions: Plasma IL-6 levels are increased in patients with cirrhosis. The severity of liver cirrhosis is an important factor for the occurrence of increased IL-6 levels. IL-6 may play a role in the hyperdynamic circulation observed in patients with cirrhosis.

Background: Colonic vascular ectasias and colorectal varices have been observed in patients with cirrhosis. However, the pathogenesis of these vascular lesions has not been established. Methods: We enrolled 35 cirrhotic patients and 20 normal controls in this study. All received colonoscopic examinations and measurements of plasma glucagon levels. Portal pressure measurements were performed in all the cirrhotic patients. Results: Colonic vascular ectasias occurred more commonly in cirrhotic patients than in controls (17 of 35 versus 0 of 20; p = 0.009) and more commonly in cirrhotic patients with ascites than in those without (15 of 24 versus 2 of 11; p = 0.038). However, the presence of colonic vascular ectasias was not related to the hepatic venous pressure gradient or plasma glucagon levels. Colorectal varices also occurred more commonly in cirrhotic patients than in controls (16 of 35 versus of 1 of 20; p = 0.034), but the hepatic venous pressure gradient, plasma glucagon levels, and severity of cirrhosis were not related to the presence of colorectal varices. Conclusion: Portal hypertension per se and increased plasma glucagon levels may not play an important role in the pathogenesis of colonic vascular ectasias or colorectal varices in patients with cirrhosis.

Background: It has been suggested that excessive formation of nitric oxide (NO) is responsible for the hyperdynamic circulation observed in portal hypertension. Substance P is a neuropeptide partly cleared by the liver and causes vasodilatation through the activation of the endothelial NO pathway. However, there are no previously published data concerning the plasma level of substance P in cirrhotic rats and its relationship to NO. Methods: Plasma concentrations of substance P and nitrate/nitrite (an index of NO production) were determined in control rats and cirrhotic rats with or without ascites using an enzyme-linked immununosorbent assay and a colorimetric assay, respectively. In addition, systemic and portal hemodynamics were evaluated by a thermodilution technique and catheterization. Results: Cirrhotic rats with and without ascites had a lower systemic vascular resistance (2.6 ± 0.2 and 3.9 ±0.4 mmHg-ml-1 min 100 g body weight, respectively) and higher portal pressure (14.6 ± 0.6 and 11.3 ± 1.8 mmHg) than control rats (6.5 ± 0.3 mmHg mP1 min 100 g BW and 6.8 ± 0.2 mmHg, respectively, P < 0.05), and cirrhotic rats with ascites had the lowest systemic vascular resistance. Plasma levels of nitrate/nitrite progressively increased in relation to the severity of liver dysfunction (control rats, 2.7 ± 0.5 nmol/ml; cirrhotic rats without ascites, 5.6 ±1.3 nmol/ml; cirrhotic rats with ascites, 8.3 ± 2.2 nmol/ml; P < 0.05). Cirrhotic rats with ascites displayed higher plasma values of substance P (57.7 ± 5.9 pg/ml) than cirrhotic rats without ascites (37.9 ± 3.1 pg/ml, P < 0.05) and control rats (30.1 ± 1.0 pg/ml, P < 0.05). There was no significant difference in plasma substance P values between control rats and cirrhotic rats without ascites (P ± 0.05). No correlation was found between plasma levels of substance P and nitrate/nitrite (r = 0.318, P ± 0.05). Conclusions: Excessive formation of NO may be responsible, at least partly, for the hemodynamic derangements in cirrhosis. Although substance P may not participate in the initiation of a hyperdynamic circulation in cirrhosis, it may contribute to the maintenance of the hyperdynamic circulation observed in cirrhotic rats with ascites.

Background: Despite many attempts to create esophageal varices in experimental animals, most of them have failed. This study investigated whether rats with partial portal vein ligation (PVL) and left adrenal vein ligation (LAL) develop hyperdynamic circulation and dilated esophageal submucosal veins as compared with sham-operated (Sham) plus LAL rats. Methods: Two series of experiments were performed to measure (a) systemic and portal hemodynamics and (b) the cross-sectional area of esophageal submucosal veins in Sham, PVL, Sham plus LAL, and PVL plus LAL rats. Hemodynamic studies with a thermodilution technique and esophageal morphometric analyses were performed 14 days after the operation. Results: PVL rats with or without LAL had a significantly lower mean arterial pressure and systemic vascular resistance accompanied by a significantly higher cardiac index and portal pressure than Sham rats with or without LAL (P < 0.05). LAL did not induce changes in mean arterial pressure, cardiac index, systemic vascular resistance, heart rate, or portal pressure in either Sham or PVL rats (P ± 0.05). The mean cross-sectional area of esophageal submucosal veins in PVL rats with LAL (7340 ± 833 pm2) was significantly larger than that in Sham rats with LAL (4236 ± 556 |inr; P <0.05). There was no significant difference in the mean cross-sectional area of esophageal submucosal veins between PVL and Sham rats without LAL. Conclusions: PVL rats with LAL developed hyperdynamic circulation similar to PVL rats without LAL. In addition, PVL plus LAL rats had larger esophageal submucosal veins than Sham plus LAL rats. This study shows that the esophageal submucosal veins of the 14-day partially portal vein-ligated rats with LAL resemble the structural abnormalities observed in human esophageal varices, suggesting that this model could be useful to investigate this entity.