- Email: [email protected]
The role of octreotide on renal function in patients with advanced cirrhosis
58
Poster Sessions
Results: At a cut-off value of 30%, RASC had 75% sensitivity, 91% specificity, 86% diagnostic accuracy, positive and negative predictive values are 81%, 88% respectively; at a cut-off value of 40 mg/dl, AFC had 92%, 95%, 86%, 73% and 95% respectively for the diagnosis of cirrhosis. SerumAscites Albumin Gradient (SAAG) were > 1.1 g/d1 in all subjects. Ascites and serum cholesterol, total protein, albumin rates were positively correlated with each other and negatively correlated with SAAG. Conclusions: These findings suggest that AFC is important in patients with PSPH. Measurements of AFC > 40 mgldl and RASC > 30% may reflect that those patients are in the precirrhotic stage of liver disease.
I 179
BASAL AND POST METHIONINE
PLASMA
HOMOCYSTEINE
IN A RAT MODEL OF CHOLESTASIS
M.R. Ebrahimkhani’, M. Dehghani’, H. Sadeghipour’, A. Shariftabrizi2, l? Pasalar2, H. Farsam3, A.R. Dehpour’. ‘Pharmacology, School Of Medicine, Tehran University Of Medical Sciences, Tehran, Iran; 2Biochemistry, School Of Medicine, Tehran University Of Medical Sciences, Tehran, Iran; ‘Pharmacy, Tehran University Of Medical Sciences, Tehran, Iran Homocysteine (Hey) has been reported as a risk factor for endothelial dysfunction and liver fibrosisTwo intersecting pathways, the methionine cycle and transsulfuration sequence compose Hey metabolism. Methionine synthase (MS) and methionine adenosyltransferase(MAT) are important enzymes in this regard. It has been demonstrated that S-nitrosylation of MAT leeds to inactivation of the enzyme. Moreover nitric oxide (NO) was mentioned as an inhibitor of MS activity. This study was aimed to investigate Hey metabolism in a rat model of cholestasis which had been shown to have increased systemic production of NO. Cholestasis was induced by bile duct ligation (BDL) and sham-operated rats were considered as controls. The animals were studied on 7th, 14th, 21st, 28th day after operation. Fasting plasma Hey and 2 hours of post-methionine load (40 and 100 mg/kg, i.p.) Were determined. Plasma nitrite/nitrate as an indicator of NO level were also measured.A significant rise in Hey level of cholestatic animals for lOOmg/kg methionine preload after 1 week was observed (P
I
180
PREVALENCE SPONTANEOUS
AND CLINICAL
RELEVANCE
PORTO-SYSTEMIC
PATIENTS WITH RECURRENT A CASE CONTROL
SHUNTS
OF IN CIRRHOTIC
HEPATIC ENCEPHALOPATHY:
STUDY
C. Efrati’,
C. Catalano2, F. Pediconi2, 0. Mecarelli3, G. Nicolini’, F. Nicolao’, l? Tanzilli’, S. Angeloni’, M. Merli’, 0. Riggio’. ‘ZZ Gastroenterology, University Of Rome ‘La Sapienza’, Rome, Italy; 2Department Of Radiology, University Of Rome ‘La Sapienza’, Rome, Italy; ‘Department Of Neurological Sciences, University Of Rome ‘La Sapienza’, Rome, Italy Introduction: Spontaneous large porto-systemic shunts may occur in cirrhotics and may be associated with hepatic encephalopathy (HE). Aim: To assess the prevalence of porto-systemic shunts in patients with recurrent HE in comparison to patients without HE, matched for age, sex and degree of liver failure. Methods: Eleven cirrhotic patients with recurrent HE (3F/8M; age: 64.1&9.3yrs; 4 alcoholic; Child-Pugh class: 4B17C; Mini-mental-state: 18.12&10.5; Trail-Making-Test Z-score: 2.2&1.6; EEG-grade Parsons-
Smith classification: lA/3B/5C/2D; PSE-index 0.48&0.04; arterial ammonia: 212.4&71.4 kg/dl) and 7 without previous or present signs of HE (7M; age: 63.1&10.6yrs; 7 viral; Child-Pugh class: 6BllC; Mini-mentalstate: 28&1.9; Trail-Making-Test Z-score: 0.5&1.6; EEG grade ParsonsSmith classification: 5 normal/l B/l C; PSE-index: 0.06&0.06; arterial ammonia: 86.3&35.8 kg/dl) were enrolled. Each patient was submitted to neurological assessment, EEG and cerebral MR in order to exclude other organic neurological pathologies. The presence of porto-systemic shunts was blindly assessed by portal venous phase multidetector-row spiral CT (MDCT). Results: The two groups were similar for age, sex, etiology of cirrhosis, and MELD score. In 8 out of 11 patients with recurrent HE (73%), CT demonstrated the presence of spontaneous porto-systemic shunts (7 large spleno-renal and one mesenteric-renal). No patient in the control group had porto-systemic shunts (x2=9.16; df=l.O; p=O.O02). The patients with HE had less frequently ascites (p=O.O07) and medium/large esophageal varices (p=O.O4). Conclusions: Our study suggests that in patients with recurrent HE, spontaneous porto-systemic shunts occur more frequently than in patients without HE. Patients with recurrent HE refractory to standard medical therapy should be studied for the presence of spontaneous shunts.
I 181
THE EFFECTS -BLOCKER
OF CARVEDILOL
A NONSELECTIVE
ON PORTAL HEMODYNAMICS
BETA
IN CIRRHOSIS
C. Fierbinteanu-Braticevici,
M. Udeanu, D. Andronescu. Gastroenterology, University Hospital, Bucharest, Romania
Portal hypertension is the result of increased hepatic resistance and portal influx. Aim: to assess the effects of Carvedilol, a third-generation nonselective bblocker with al- adrenergic activity on portal and systemic homodynamic in patients with cirrhosis and portal hypertension. Methods: Fifty patients with cirrhosis and portal hypertension were divided in two groups and statistically compared as follows: group I- 25 patients received Carvedilol, 12, 5 mg/day and group II- 25 patients received placebo for six days. All the patients had hemodynamic, endocrine and renal measurements before and after administration of Carvedilol or placebo. They underwent inulin clearance, lithium clearance, plasma rennin activity, concentration of plasma aldosteron and urinary sodium excretion. Hemodynamic effects were assessed by portal flow volume and velocity, cardiac output, medium blood pressure and hepatic venous pressure gradient (HVPG). Results: Carvedilol markedly reduced the HVPG (~‘0, 001) and increased the portal blood flow and velocity (~‘0, 05). Carvedilol reduced the medium blood pressure (pi 0, 001) with statistically insignificant alterations in inulin clearance and 24.hours urinary sodium excretion. Carvedilol also reduced the concentration of plasma aldosteron (p < 0, 002). Conclusions: Carvedilol can be used as an alternative drug for the prophylactic treatment of portal hypertension with careful monitoring of blood pressure regarding its hypotensive effects.
I 182
THE ROLE OF OCTREOTIDE
ON RENAL FUNCTION
PATIENTS WITH ADVANCED
CIRRHOSIS
IN
C. Fierbinteanu-Braticevici,
A. Bengus, D. Andronescu. Gastroenterology, University Hospital, Bucharest, Romania
Octreotide the synthetic somatostatin analog has replaced vasopressin in the treatment of bleeding esophageal varices. Aim: to assess the effects of ocreotide on renal function in cirrhotic patients with portal hypertension. Methods: Forty six patients with advanced cirrhosis were divided in two groups and statistically compared. Group I -23 patients, received 50 mg intravenous octreotide per hour for three days. Group II - 23 patients
Category 2: Cirrhosis and its complications, received placebo. All the patients underwent before and after 3 days of treatment, renal function tests which consist of: serum creatinine, urinary sodium concentration, inulin clearance as index of glomerular filtration rate (GFR), wather dim-e&, lithium clearance, plasma renin activity and aldosteron concentration. Systemic hemodinamymics were assessed by medium blood pressure and cardiac output. Results: Octreotide produced an impressive improvement in renal function: improved inulin clearance (p
I
183
OCTREOTIDE INDUCED
ANTAGONIZES
THE UROTENSIN
AORTIC CONTRACTIONS
CIRRHOTIC
59
per hour for five days). We measured the hepatic venous pressure gradient (HVPG) after endoscopy therapy within the first 24 hours after admission. Results: Eleven patients died (16%) and 56 did not died (84%) at 6-week. Cox regression analysis showed that HVPG, Child-Pugh score, and treatment failure were independent predictors of survival. HVPG gradient of at least 20 mmHg identify patients with increased risk of dying. Nine of 26 patients who had this HVPG value died vs. 2 of 41 patients who did not died, odds ratio 10. 32 (95% confidence interval, 2. O-52. 9; P=O. 002). Multivariate analysis identified HVPG as the only independent variable predicting treatment failure (failure to control bleeding or early variceal rebleeding), which occurred in 14 (54%) and 5 in patients (12%) with HVPG above and below 20 mmHg, respectively (P=O. 0002). In addition, patients with HVPG of 20 mmHg had a significantly longer intensive care unit stay, longer hospital stay, greater blood transfusions. Conclusions: In cirrhotic patients with variceal bleeding, early measurement of HVPG may predict survival and identify patients with high risk of failure to treatment.
AND
I 185
RATS
M. Schepke, M. Neef, T. Sauerbruch. Department Of General Internal Medicine, University Of Bonn, Bonn, Germany
Introduction and Aim: U II plasma levels are increased in patients with cirrhosis and correlate with the degree of portal hypertension (.I Hepato1 2002;37: 767-772). The vasoactive peptide Urotensin II has structural similarities to octreotide. Octreotide is used in the treatment of variceal bleeding and improves in vitro the reduced vascular contractility to c1adrenoceptoragonists in rats with cirrhosis (Gastroenterology 2001;120: 975-983). The aim of our study was to investigate whether octreotide interferes with the U II induced aortic contractions of rats with secondary biliary cirrhosis and shamoperated controls. Methods: U II induced (10~4-10~7 mol/l) isometric contractions of endothelium free aortic rings were determined in 6 rats with secondary biliary cirrhosis and 6 shamoperated controls. At the maximal contraction octreotide was added cumulatively to the organ bath lo-‘-3 x 10m6 mol/l). Results: Maximal U II induced contraction of endothelium free aortic rings were significantly decreased in cirrhotic rats compared to shamoperated rats (Emax: cirrhosis mean0.6&SEMO.l g; sham l.O&O.l g; p
EARLY MEASUREMENTS
OF PORTAL PRESSURE
SURVIVAL
PATIENTS WITH ACUTE VARICEAL
IN CIRRHOTIC
and clinical aspects
II (U II)
IN NORMAL
J. Heller, N. Fischer,
I 184
pathophysiology
PREDICTS
HEMORRHAGE
L. Ruiz-del-Arbol’, A. Monescillo2, M. R. Gonzalez’, .I. M. Urman’, M. Gonzalez-Garcia’, E. Pe a’, A. Cane’, V. Moreira’. ’Gastroenterology, Hospital Ramon Y Cajal, Madrid, Spain; 2Gastroenterology, Hospital Insular De Gran Canaria, Canary Islands, Spain Background: Bleeding from esophageal varices is a major cause of mortality in patients with cirrhosis and portal hypertension. However, whether measurements of portal pressure predict the survival of patients with acute variceal hemorrhage has not been investigated. Methods: In a prospective study, 67 patients with cirrhosis and acute variceal bleeding underwent emergency banding ligation plus somatostatin (a 250.kg intravenous bolus followed by an infusion at a rate of 250 kg
ENDOSCOPIC
TREATMENT
WITH ARGON
PLASMA
COAGULATION
FOR PORTAL HYPERTENSIVE
GASTROPATHY
WITH NO RESPONSE
PHARMACOLOGICAL
TO
THERAPY
B. Gonzalez- Suarez, D. Monfort, M. Piqueras, M. Planella, C. Aracil, A. Gallego, X. Ton-as, J.M. Lopez-Balaguer, C. Villanueva, J. Balanzo. Liver Unit, Sant Pau Hospital, Barcelona, Spain Portal hypertensive gastropathy (PHG) is a rare but relevant cause of gastrointestinal bleeding for which few options of medical therapy are available. Argon plasma coagulation (APC) is an endoscopic non-contact method of electrocoagulation with a limited deep of penetration and which may be applied tangentially. It may be used to treat widespread vascular disorders as PHG. The aim of this study was to assess the efficacy of APC for the treatment of chronic bleeding from PHG with no response to pharmacological therapy of portal hypertension. Methods: 22 cirrhotic patients (Pugh class A/B/C in 1111011) were included. 16 patients had PHG with chronic anemia despite treatment with nadolol + isosorbide mononitrate and 6 with acute bleeding episode. PHG was diffuse in 9 patients and was mainly antral in 13. APC was delivered by multiple and brief pulses applied over wide areas of visible angioectasias. Sessions were performed every 3 weeks. Results: A mean of 5.8 sessions of APC were performed during a period of 9 + 7 months. The mean follow-up was 36 + 24 months. Hemoglobin value significantly improved after APC (from 75+12 g/l to 106+17 g/l, P
Poster Sessions
Results: At a cut-off value of 30%, RASC had 75% sensitivity, 91% specificity, 86% diagnostic accuracy, positive and negative predictive values are 81%, 88% respectively; at a cut-off value of 40 mg/dl, AFC had 92%, 95%, 86%, 73% and 95% respectively for the diagnosis of cirrhosis. SerumAscites Albumin Gradient (SAAG) were > 1.1 g/d1 in all subjects. Ascites and serum cholesterol, total protein, albumin rates were positively correlated with each other and negatively correlated with SAAG. Conclusions: These findings suggest that AFC is important in patients with PSPH. Measurements of AFC > 40 mgldl and RASC > 30% may reflect that those patients are in the precirrhotic stage of liver disease.
I 179
BASAL AND POST METHIONINE
PLASMA
HOMOCYSTEINE
IN A RAT MODEL OF CHOLESTASIS
M.R. Ebrahimkhani’, M. Dehghani’, H. Sadeghipour’, A. Shariftabrizi2, l? Pasalar2, H. Farsam3, A.R. Dehpour’. ‘Pharmacology, School Of Medicine, Tehran University Of Medical Sciences, Tehran, Iran; 2Biochemistry, School Of Medicine, Tehran University Of Medical Sciences, Tehran, Iran; ‘Pharmacy, Tehran University Of Medical Sciences, Tehran, Iran Homocysteine (Hey) has been reported as a risk factor for endothelial dysfunction and liver fibrosisTwo intersecting pathways, the methionine cycle and transsulfuration sequence compose Hey metabolism. Methionine synthase (MS) and methionine adenosyltransferase(MAT) are important enzymes in this regard. It has been demonstrated that S-nitrosylation of MAT leeds to inactivation of the enzyme. Moreover nitric oxide (NO) was mentioned as an inhibitor of MS activity. This study was aimed to investigate Hey metabolism in a rat model of cholestasis which had been shown to have increased systemic production of NO. Cholestasis was induced by bile duct ligation (BDL) and sham-operated rats were considered as controls. The animals were studied on 7th, 14th, 21st, 28th day after operation. Fasting plasma Hey and 2 hours of post-methionine load (40 and 100 mg/kg, i.p.) Were determined. Plasma nitrite/nitrate as an indicator of NO level were also measured.A significant rise in Hey level of cholestatic animals for lOOmg/kg methionine preload after 1 week was observed (P
I
180
PREVALENCE SPONTANEOUS
AND CLINICAL
RELEVANCE
PORTO-SYSTEMIC
PATIENTS WITH RECURRENT A CASE CONTROL
SHUNTS
OF IN CIRRHOTIC
HEPATIC ENCEPHALOPATHY:
STUDY
C. Efrati’,
C. Catalano2, F. Pediconi2, 0. Mecarelli3, G. Nicolini’, F. Nicolao’, l? Tanzilli’, S. Angeloni’, M. Merli’, 0. Riggio’. ‘ZZ Gastroenterology, University Of Rome ‘La Sapienza’, Rome, Italy; 2Department Of Radiology, University Of Rome ‘La Sapienza’, Rome, Italy; ‘Department Of Neurological Sciences, University Of Rome ‘La Sapienza’, Rome, Italy Introduction: Spontaneous large porto-systemic shunts may occur in cirrhotics and may be associated with hepatic encephalopathy (HE). Aim: To assess the prevalence of porto-systemic shunts in patients with recurrent HE in comparison to patients without HE, matched for age, sex and degree of liver failure. Methods: Eleven cirrhotic patients with recurrent HE (3F/8M; age: 64.1&9.3yrs; 4 alcoholic; Child-Pugh class: 4B17C; Mini-mental-state: 18.12&10.5; Trail-Making-Test Z-score: 2.2&1.6; EEG-grade Parsons-
Smith classification: lA/3B/5C/2D; PSE-index 0.48&0.04; arterial ammonia: 212.4&71.4 kg/dl) and 7 without previous or present signs of HE (7M; age: 63.1&10.6yrs; 7 viral; Child-Pugh class: 6BllC; Mini-mentalstate: 28&1.9; Trail-Making-Test Z-score: 0.5&1.6; EEG grade ParsonsSmith classification: 5 normal/l B/l C; PSE-index: 0.06&0.06; arterial ammonia: 86.3&35.8 kg/dl) were enrolled. Each patient was submitted to neurological assessment, EEG and cerebral MR in order to exclude other organic neurological pathologies. The presence of porto-systemic shunts was blindly assessed by portal venous phase multidetector-row spiral CT (MDCT). Results: The two groups were similar for age, sex, etiology of cirrhosis, and MELD score. In 8 out of 11 patients with recurrent HE (73%), CT demonstrated the presence of spontaneous porto-systemic shunts (7 large spleno-renal and one mesenteric-renal). No patient in the control group had porto-systemic shunts (x2=9.16; df=l.O; p=O.O02). The patients with HE had less frequently ascites (p=O.O07) and medium/large esophageal varices (p=O.O4). Conclusions: Our study suggests that in patients with recurrent HE, spontaneous porto-systemic shunts occur more frequently than in patients without HE. Patients with recurrent HE refractory to standard medical therapy should be studied for the presence of spontaneous shunts.
I 181
THE EFFECTS -BLOCKER
OF CARVEDILOL
A NONSELECTIVE
ON PORTAL HEMODYNAMICS
BETA
IN CIRRHOSIS
C. Fierbinteanu-Braticevici,
M. Udeanu, D. Andronescu. Gastroenterology, University Hospital, Bucharest, Romania
Portal hypertension is the result of increased hepatic resistance and portal influx. Aim: to assess the effects of Carvedilol, a third-generation nonselective bblocker with al- adrenergic activity on portal and systemic homodynamic in patients with cirrhosis and portal hypertension. Methods: Fifty patients with cirrhosis and portal hypertension were divided in two groups and statistically compared as follows: group I- 25 patients received Carvedilol, 12, 5 mg/day and group II- 25 patients received placebo for six days. All the patients had hemodynamic, endocrine and renal measurements before and after administration of Carvedilol or placebo. They underwent inulin clearance, lithium clearance, plasma rennin activity, concentration of plasma aldosteron and urinary sodium excretion. Hemodynamic effects were assessed by portal flow volume and velocity, cardiac output, medium blood pressure and hepatic venous pressure gradient (HVPG). Results: Carvedilol markedly reduced the HVPG (~‘0, 001) and increased the portal blood flow and velocity (~‘0, 05). Carvedilol reduced the medium blood pressure (pi 0, 001) with statistically insignificant alterations in inulin clearance and 24.hours urinary sodium excretion. Carvedilol also reduced the concentration of plasma aldosteron (p < 0, 002). Conclusions: Carvedilol can be used as an alternative drug for the prophylactic treatment of portal hypertension with careful monitoring of blood pressure regarding its hypotensive effects.
I 182
THE ROLE OF OCTREOTIDE
ON RENAL FUNCTION
PATIENTS WITH ADVANCED
CIRRHOSIS
IN
C. Fierbinteanu-Braticevici,
A. Bengus, D. Andronescu. Gastroenterology, University Hospital, Bucharest, Romania
Octreotide the synthetic somatostatin analog has replaced vasopressin in the treatment of bleeding esophageal varices. Aim: to assess the effects of ocreotide on renal function in cirrhotic patients with portal hypertension. Methods: Forty six patients with advanced cirrhosis were divided in two groups and statistically compared. Group I -23 patients, received 50 mg intravenous octreotide per hour for three days. Group II - 23 patients
Category 2: Cirrhosis and its complications, received placebo. All the patients underwent before and after 3 days of treatment, renal function tests which consist of: serum creatinine, urinary sodium concentration, inulin clearance as index of glomerular filtration rate (GFR), wather dim-e&, lithium clearance, plasma renin activity and aldosteron concentration. Systemic hemodinamymics were assessed by medium blood pressure and cardiac output. Results: Octreotide produced an impressive improvement in renal function: improved inulin clearance (p
I
183
OCTREOTIDE INDUCED
ANTAGONIZES
THE UROTENSIN
AORTIC CONTRACTIONS
CIRRHOTIC
59
per hour for five days). We measured the hepatic venous pressure gradient (HVPG) after endoscopy therapy within the first 24 hours after admission. Results: Eleven patients died (16%) and 56 did not died (84%) at 6-week. Cox regression analysis showed that HVPG, Child-Pugh score, and treatment failure were independent predictors of survival. HVPG gradient of at least 20 mmHg identify patients with increased risk of dying. Nine of 26 patients who had this HVPG value died vs. 2 of 41 patients who did not died, odds ratio 10. 32 (95% confidence interval, 2. O-52. 9; P=O. 002). Multivariate analysis identified HVPG as the only independent variable predicting treatment failure (failure to control bleeding or early variceal rebleeding), which occurred in 14 (54%) and 5 in patients (12%) with HVPG above and below 20 mmHg, respectively (P=O. 0002). In addition, patients with HVPG of 20 mmHg had a significantly longer intensive care unit stay, longer hospital stay, greater blood transfusions. Conclusions: In cirrhotic patients with variceal bleeding, early measurement of HVPG may predict survival and identify patients with high risk of failure to treatment.
AND
I 185
RATS
M. Schepke, M. Neef, T. Sauerbruch. Department Of General Internal Medicine, University Of Bonn, Bonn, Germany
Introduction and Aim: U II plasma levels are increased in patients with cirrhosis and correlate with the degree of portal hypertension (.I Hepato1 2002;37: 767-772). The vasoactive peptide Urotensin II has structural similarities to octreotide. Octreotide is used in the treatment of variceal bleeding and improves in vitro the reduced vascular contractility to c1adrenoceptoragonists in rats with cirrhosis (Gastroenterology 2001;120: 975-983). The aim of our study was to investigate whether octreotide interferes with the U II induced aortic contractions of rats with secondary biliary cirrhosis and shamoperated controls. Methods: U II induced (10~4-10~7 mol/l) isometric contractions of endothelium free aortic rings were determined in 6 rats with secondary biliary cirrhosis and 6 shamoperated controls. At the maximal contraction octreotide was added cumulatively to the organ bath lo-‘-3 x 10m6 mol/l). Results: Maximal U II induced contraction of endothelium free aortic rings were significantly decreased in cirrhotic rats compared to shamoperated rats (Emax: cirrhosis mean0.6&SEMO.l g; sham l.O&O.l g; p
EARLY MEASUREMENTS
OF PORTAL PRESSURE
SURVIVAL
PATIENTS WITH ACUTE VARICEAL
IN CIRRHOTIC
and clinical aspects
II (U II)
IN NORMAL
J. Heller, N. Fischer,
I 184
pathophysiology
PREDICTS
HEMORRHAGE
L. Ruiz-del-Arbol’, A. Monescillo2, M. R. Gonzalez’, .I. M. Urman’, M. Gonzalez-Garcia’, E. Pe a’, A. Cane’, V. Moreira’. ’Gastroenterology, Hospital Ramon Y Cajal, Madrid, Spain; 2Gastroenterology, Hospital Insular De Gran Canaria, Canary Islands, Spain Background: Bleeding from esophageal varices is a major cause of mortality in patients with cirrhosis and portal hypertension. However, whether measurements of portal pressure predict the survival of patients with acute variceal hemorrhage has not been investigated. Methods: In a prospective study, 67 patients with cirrhosis and acute variceal bleeding underwent emergency banding ligation plus somatostatin (a 250.kg intravenous bolus followed by an infusion at a rate of 250 kg
ENDOSCOPIC
TREATMENT
WITH ARGON
PLASMA
COAGULATION
FOR PORTAL HYPERTENSIVE
GASTROPATHY
WITH NO RESPONSE
PHARMACOLOGICAL
TO
THERAPY
B. Gonzalez- Suarez, D. Monfort, M. Piqueras, M. Planella, C. Aracil, A. Gallego, X. Ton-as, J.M. Lopez-Balaguer, C. Villanueva, J. Balanzo. Liver Unit, Sant Pau Hospital, Barcelona, Spain Portal hypertensive gastropathy (PHG) is a rare but relevant cause of gastrointestinal bleeding for which few options of medical therapy are available. Argon plasma coagulation (APC) is an endoscopic non-contact method of electrocoagulation with a limited deep of penetration and which may be applied tangentially. It may be used to treat widespread vascular disorders as PHG. The aim of this study was to assess the efficacy of APC for the treatment of chronic bleeding from PHG with no response to pharmacological therapy of portal hypertension. Methods: 22 cirrhotic patients (Pugh class A/B/C in 1111011) were included. 16 patients had PHG with chronic anemia despite treatment with nadolol + isosorbide mononitrate and 6 with acute bleeding episode. PHG was diffuse in 9 patients and was mainly antral in 13. APC was delivered by multiple and brief pulses applied over wide areas of visible angioectasias. Sessions were performed every 3 weeks. Results: A mean of 5.8 sessions of APC were performed during a period of 9 + 7 months. The mean follow-up was 36 + 24 months. Hemoglobin value significantly improved after APC (from 75+12 g/l to 106+17 g/l, P