Jaime Bosch

Hepatic Expression of CXC Chemokines Predicts Portal Hypertension and Survival in Patients With Alcoholic Hepatitis

BACKGROUND & AIMS Alcoholic hepatitis (AH) is characterized by hepatocellular damage, inflammation, and fibrosis. We performed a prospective study to associate hepatic expression of the CXC subfamily of chemokines with histology findings and prognosis of patients with AH. METHODS Liver biopsy samples from 105 patients with AH and 5 normal liver samples (controls) were evaluated for steatosis, inflammation, fibrosis, and cholestasis. Computer-based morphometric analysis assessed the numbers of infiltrating CD3+ T cells and CD15+ cells (neutrophils); terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling staining was used to quantify apoptosis. Expression of CXC and CC chemokines and selected signaling components were assessed by quantitative reverse-transcription polymerase chain reaction; protein levels of interleukin (IL)-8 and Gro-alpha also were determined by immunohistochemistry. Serum levels of IL-8 and Gro-alpha were measured by enzyme-linked immunosorbent assay. The Cox regression model identified variables associated with mortality. RESULTS Most patients (75%) had severe AH; their 90-day mortality rate was 21.9%. In AH liver samples, expression of the CXC subfamily members IL-8, Gro-alpha, CXCL5, CXCL6, CXCL10, and platelet factor 4 was up-regulated and compared with controls. The CC chemokine CCL2, but not CCL5, also was up-regulated. Higher expression levels of IL-8, CXCL5, Gro-gamma, and CXCL6 were associated with worse prognosis. Expression of CXC components correlated with neutrophil infiltration and the severity of portal hypertension. In the multivariate analysis, IL-8 protein levels were an independent predictor of 90-day mortality. IL-8 and Gro-alpha serum levels did not correlate with prognosis. CONCLUSIONS Hepatic expression of CXC components correlates with prognosis of patients with AH. Reagents that target CXC chemokines might be developed as therapeutics.

Von Willebrand factor levels predict clinical outcome in patients with cirrhosis and portal hypertension

Background and aims Endothelial dysfunction is a major determinant of the increased hepatic vascular tone of cirrhotic livers. Von Willebrand factor (vWF), P-selectin and 8-iso-PGF2α (isoprostanes), surrogate markers of endothelial dysfunction, are increased in patients with cirrhosis. This study was aimed at exploring in patients with cirrhosis and portal hypertension the relation of these endothelial factors with systemic and hepatic haemodynamics and their possible clinical prognostic value. Methods 42 consecutive patients with cirrhosis and portal hypertension had measurement of the hepatic venous pressure gradient (HVPG), cardiopulmonary pressures and vWF, P-selectin and isoprostane levels in blood samples from hepatic and peripheral veins. Patients were followed up to 2 years, death or liver transplantation and any clinical event were recorded. Results vWF, P-selectin and isoprostanes were increased in patients with cirrhosis compared with controls (p<0.001). vWF levels significantly correlated with HVPG, Child–Pugh score and MELD. Cox model analysis disclosed an independent indirect association of peripheral vWF with survival free of portal hypertension-related events and of transplantation. The vWF cut-off value of 216 U/dl (Youden index) disclosed two different populations of patients with cirrhosis with a highly different probability of survival free of portal hypertension-related events and transplantation (87% vs 22%, p=0.001). The prognostic role of vWF persisted after adjusting for parameters of liver dysfunction and for HVPG. Conclusions In patients with cirrhosis and portal hypertension vWF levels correlate with liver function and HVPG and independently predict clinical outcome.

Patients Whose First Episode of Bleeding Occurs While Taking a β-Blocker Have High Long-term Risks of Rebleeding and Death

BACKGROUND & AIMS Patients who have their first episode of variceal bleeding despite primary prophylaxis with a nonselective β-adrenergic receptor antagonist (also called a nonselective β-blocker [NSBB]) receive additional treatment by endoscopic band ligation to prevent further bleeding. However, little is known about their long-term outcomes. METHODS We collected data on 89 consecutive patients with cirrhosis who were admitted to the Liver Unit of Hospital Clínic, Barcelona, with acute esophageal variceal bleeding between June 2007 and February 2011. Thirty-four patients were receiving primary prophylaxis with NSBBs when they had their first episode of variceal bleeding, whereas 55 were not receiving NSBBs (controls). All patients were subsequently treated with a combination of endoscopic band ligation and NSBBs. Patients were examined after 1, 3, and 6 months and every 6 months thereafter until 2 years. RESULTS After 2 years, a greater proportion of patients who had their first episode of bleeding while on NSBBs had further bleeding, compared with controls (48% vs 24%; P = .01). Primary prophylaxis with NSBBs and serum levels of bilirubin were independent predictors of rebleeding. Overall, 11 patients died, and 5 underwent liver transplantation. Liver transplantation-free survival was lower among patients who had their first episode of bleeding while taking NSBBs (66% vs 88% for controls; P = .02). Primary prophylaxis with NSBBs and Child-Pugh class were independently associated with liver transplantation-free survival. CONCLUSIONS Patients who have their first episode of variceal bleeding while on primary prophylaxis with a β-blocking agent have an increased risk of further bleeding and death, despite adding endoscopic band ligation. These patients possibly require alternative treatment approaches.

Pathophysiology of variceal bleeding

Variceal formation is an almost unavoidable complication of cirrhosis. Varices are already present in about 40% of compensated, asymptomatic patients at diagnosis, and the incidence increases to up to 90% of patients on long-term follow up1. The annual incidence is around 6% per year2.

Possibilities of manipulating nitric oxide biosynthesis in the treatment of portal hypertension: statins

Statins are lipid-lowering agents that act by inhibiting the activity of the ratelimiting enzyme for cholesterol synthesis, the 3-hydroxyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase, Statins are among the most widely prescribed drug class in Western countries, and have been consistently shown to decrease the incidence of cardiovascular events and to improve survival in patients with ischemic heart disease1. Although the beneficial effects of statins were initially entirely attributed to their lipid-lowering effects, detailed analysis of randomized controlled trials suggested that some of their benefits were independent of their cholesterol-lowering capacity2, 3. This prompted further investigations that showed a variety of effects of statins beyond cholesterol reduction, which have been called the pleiotropic effects of statins4, Statins reduce oxidant stress and inflammation at the vessel wall, have anti thrombotic and anti proliferative properties and improve endothelial function, increasing nitric oxide (NO) production in endothelial cells5, 6, 7, 8, 9, 10, 11, Statins’ effects on angiogenesis are more controversial, but it seems that at clinically relevant doses statins are proangiogenic agents 7, 12, 13. Many of these effects have been demonstrated to occur as early as within 24 hours of statin administration, further reinforcing their independence from plasma cholesterol changes14, 15, 16.

What else we need

Available evidence supports the concept that repeat measurements of hepatic venous pressure gradient (HVPG) adequately predict the effectiveness of drug therapy for portal hypertension1, 2, 3, 4, 5, 6, 7. The HVPG response to continued pharmacologic therapy correlates well with the risk of first variceal bleeding3, 8 or rebleeding2, 4. Patients who experienced a decrease in HVPG to < 12 mmHg have an extremely low risk of rebleeding (“optimal” hemodynamic response). Even if the final HVPG is not reduced < 12 mmHg, a decrease of over 20% of baseline values is also associated with a marked reduction in the risk of bleeding of about 19.5% vs 45% in non-responders (“good” hemodynamic response) (Table 1). Unfortunately such hemodynamic responses are achieved in only about 30–60% of patients receiving non-selective beta-blockers. Lack of achievement of these targets constitutes the strongest independent predictor of variceal bleeding or rebleeding2, 3, 4, 10. Therefore, ideally, the treatment of portal hypertension should be guided by measuring the individual portal pressure response to therapy. This allows us to predict whether the treatment is likely to offer optimal or adequate protection from the risk of bleeding on follow-up10, 11.

The Treatment and Prevention of Variceal Bleeding

Variceal bleeding is one of the most common and severe complications of cirrhosis. Even with the current best medical care, mortality from variceal bleeding is still around 15–20%. Moreover, variceal bleeding often leads to deterioration in liver function, and it is a common trigger for other complications of cirrhosis, such as bacterial infections or hepatorenal syndrome.

Physiology of the Splanchnic and Hepatic Circulations

The splanchnic and hepatic circulations account for over a quarter of cardiac output and are highly specialized to meet the demands of these complex and complementary organs. The hepatic circulation is fundamental to liver function, and the splanchnic circulation participates in the control of systemic hemodynamics as well as perfusion of the intestine.

Endoscopic Assessment of Portal Hypertension Including Variceal Pressure Measurements

Portal hypertension is a common complication in liver diseases that should be searched for in all cirrhotic patients, as well as in other chronic liver diseases (1). Its main clinical consequence is bleeding from ruptured esophageal or gastric varices, which constitutes the major cause of death and of liver transplantation in patients with cirrhosis, the leading etiology for portal hypertension.