Search

The association of prazosin to propranolol enhances the decrease in portal pressure but may cause hypotension and sodium retention. The aim of this study was to compare the portal pressure reduction and safety of the combination of propranolol plus prazosin with that of propranolol plus isosorbide-5-mononitrate (ISMN).Fifty-six portal-hypertensive cirrhotics received randomly propranolol plus prazosin (n = 28) or propranolol plus ISMN (n = 28) orally for 3 months. Hemodynamics and liver and renal function were assessed at baseline and after 3 months.Propranolol plus prazosin caused a greater reduction in hepatic venous pressure gradient (HVPG) than propranolol plus ISMN (-24.2% +/- 11% vs. -16.1% +/- 11%; P0.01). A reduction in HVPG of20% was significantly more frequent in the propranolol plus prazosin group than in the propranolol plus ISMN group (85% vs. 53%; P0.05). Neither treatment modified hepatic blood flow, quantitative liver function test results, glomerular filtration rate, plasma renin activity, or plasma aldosterone level. Side effects occurred in 13 patients receiving propranolol plus prazosin compared with 7 receiving propranolol plus ISMN (P = 0.16).Propranolol plus prazosin has a greater portal pressure-lowering effect than propranolol plus ISMN. Both therapies were safe for liver and renal function. However, the combination of propranolol plus prazosin caused a greater decrease in arterial pressure and was less well tolerated than propranolol plus ISMN.

Background/Aims: Little information is available on hepatic arterial hemodynamics in cirrhosis because of the invasiveness of methods. Hepatic arterial resistance indexes were evaluated noninvasively by Doppler ultrasonography and were correlated with portal hemodynamics evaluated both noninvasively and invasively. Methods: Hepatic arterial resistance indexes, portal blood flow velocity and volume, and portal vein congestion index were evaluated in 31 controls and 171 cirrhotic patients with (n = 13) or without (n = 158) portal vein thrombosis. Resistance to portal blood flow was also calculated in 15 patients from hepatic venous pressure gradient, measured by hepatic vein catheterization, and portal blood flow. Results: Resistance indexes were significantly higher in cirrhotics without portal thrombosis than in controls (pulsatility index, 1.30 ± 0.29 vs. 0.89 ± 0.09; P < 0.001; resistive index, 0.71 ± 0.07 vs. 0.59 ± 0.04; P < 0.001). In patients with portal thrombosis, the pulsatility index (1.86 ± 0.39) and resistive index (0.81 ± 0.06) were significantly higher than in controls (P < 0.001) and in patients without thrombosis (P < 0.001). Resistance indexes directly correlated with portal resistance (P < 0.01), the congestion index (P < 0.01), and the degree of esophageal varices (P < 0.01). Conclusions: Hepatic arterial resistance indexes increase in cirrhosis, particularly with portal vein thrombosis. The pathophysiology of the increase in hepatic arterial resistance seems to be parallel to that of portal resistance.

Background/Aims: Transjugular intrahepatic portosystemic shunt (TIPS) placement has been used for the treatment of recurrent variceal hemorrhage. The 1-year incidence of shunt stenosis or occlusion after TIPS placement was prospectively assessed, and the accuracy of Doppler ultrasonography to predict TIPS stenosis was evaluated. Methods: Twenty-two patients with recurrent variceal hemorrhage were selected for TIPS placement between April 1991 and May 1992. Preoperative and postoperative evaluation included clinical assessment, upper gastrointestinal endoscopy, portal angiography with pressure measurements, and Doppler ultrasonography. Follow-up was performed at 3 and 12 months post-TIPS and when patients developed recurrent bleeding. Results: Twenty-one of 22 patients (Child-Pugh class A-1, B-11, C-9) had successful TIPS placement. Seventeen of 21 patients have completed follow-up for at least 12 months. Of these 17 patients, 2 of 17 (12%) developed TIPS occlusion, 7 of 17 (41%) developed shunt stenosis, and 8 of 17 (47%) showed no stenosis on follow-up angiography. Doppler ultrasonographic assessment of the TIPS predicted shunt stenosis or occlusion with 100% sensitivity, 98% specificity, and 90% positive predictive value. Conclusions: Shunt occlusion or stenosis develops frequently within 12 months after TIPS placement, and Doppler ultrasonography is accurate in the noninvasive assessment of shunt stenosis. TIPS placement without careful follow-up and shunt revision cannot be considered a long-term treatment of variceal hemorrhage.

Clinical and anamnestic data, Pugh score, and size of esophageal varices were obtained in 129 cirrhotics. Hepatic vein catheterization was performed to measure hepatic venous pressure gradient (HVPG), indocyanine green (ICG) intrinsic hepatic clearance, and hepatic plasma flow. During a follow-up period of up to 60 months, 44 patients experienced gastrointestinal bleeding and 54 died. Applying Cox regression analysis, ICG intrinsic hepatic clearance, Pugh score, previous variceal bleeding, and HVPG were the only significant prognostic determinants of survival. In addition, Cox's regression analysis showed that HVPG, Pugh score, size of varices, and previous variceal bleeding all contained significant prognostic information regarding risk of gastrointestinal bleeding. The models were validated using a split-sample technique, and prognostic indexes for death and gastrointestinal bleeding were calculated. The prognostic index predicting death had significantly improved prognostic accuracy over a prognostic index calculated excluding the data obtained from hepatic vein catheterization (P less than 0.05). In conclusion, prognostic accuracy in cirrhosis with portal hypertension is significantly improved by information obtained from hepatic vein catheterization.

Background & Aims: A reduction of the hepatic venous pressure gradient (HVPG) to ≤12 mm Hg or by ≥20% of baseline prevents variceal bleeding in cirrhosis. Because some inconsistent data have argued against the clinical application of these hemodynamic targets, we performed a systematic review of available studies from the Cochrane Library and MEDLINE. Methods: Hemodynamic targets were HVPG reduction (1) to ≤12 mm Hg; (2) by ≥20% with final value >12 mm Hg; (3) by ≥20% or to ≤12 mm Hg. Meta-regression analysis was used to explore heterogeneity. Results: Twelve studies were identified including 943 patients. Pooled odds ratios for bleeding for the 3 hemodynamic targets were, respectively, 0.21 (95% CI: 0.10–0.45; P = .0001), 0.25 (95% CI: 0.11–0.56; P = .001), and 0.17 (95% CI: 0.09–0.33; P = .001). A significant heterogeneity was found for the 2 last estimates, and meta-regression analysis showed that this was caused by an exceedingly long interval between HVPG measurements in 1 study. After exclusion of that study, heterogeneity disappeared, and the pooled odds ratios were, respectively, 0.19 (95% CI: 0.11–0.34; P = .0001) and 0.14 (95% CI: 0.09–0.21; P = .0001). The beneficial effect of HVPG reduction for first bleeding was similar to that for recurrent bleeding. Mortality was significantly reduced for HVPG reduction by ≥20% or to ≤12 mm Hg (pooled odds ratio, 0.39; 95% CI: 0.19–0.81, P = .012). Conclusions: HVPG reduction to ≤12 mm Hg or by ≥20% significantly reduces the risk of bleeding, and a reduction of ≥20% significantly reduces mortality. These hemodynamic targets should be considered for clinical practice and for randomized controlled trials.

Postprandial increases in portal pressure may influence esophageal variceal rupture. The effects of chronic propranolol and octreotide (100 and 200 microg subcutaneously in a single dose) on postprandial hemodynamics were evaluated.FIRST STUDY: 36 cirrhotic patients were studied at baseline and 30 and 60 minutes after a standard meal and then treated with propranolol (139 +/- 9 mg/d during 39 +/- 2 days). SECOND STUDY: After baseline measurements, patients were randomized into 3 groups: (1) placebo, (2) octreotide (100 microg), or (3) octreotide (200 microg) (n = 12 for each group). Thirty minutes postinjection a new baseline was established and measurements were repeated 30 and 60 minutes after the meal.First study: Baseline portal pressure was 18.1 +/- 1.2 mm Hg, 30 and 60 minutes after the meal it was 21.5 +/- 0.8 mm Hg and 20.5 +/- 0.8 mm Hg, respectively (both P0.01 vs. baseline). Cardiac index (CI) was 4.5 +/- 0.2, 4.8 +/- 0.2, and 4.9 +/- 0.2 L x min(-1) x m(-2), respectively (both P0.05 vs. baseline). Peripheral vascular resistance was 1012 +/- 56, 902 +/- 51 (P = NS), and 884 +/- 49 dynes x sec x cm(-5) (P0.05 vs. baseline), respectively. Second study: Propranolol and placebo did not blunt postprandial increase in portal pressure. Octreotide (100 microg) partially ameliorated postprandial increase in portal pressure. Octreotide (200 microg) significantly enhanced the portal hypotensive effect of propranolol and blunted the postprandial increase in portal pressure.Octreotide blunts postprandial increase in portal pressure not prevented by long-term propranolol administration.

Background & Aims: A number of transporters and channels have been identified in cholangiocytes, but the role that bile ducts play in the formation of bile in vivo is unclear.We determined the contribution of cholangiocytes to bile flow and biliary bicarbonate excretion in normal rat liver. Methods: Bile flow and biliary bicarbonate were measured in isolated rat livers perfused via both the portal vein and the hepatic artery because the hepatic artery provides the blood supply to bile ducts.Livers were perfused with secretin or acetylcholine (ACh), which respectively increase either adenosine 3',5'-cyclic monophosphate (cAMP) or cytosolic Ca 2+ in cholangiocytes.Livers also were perfused with glucagon or vasopressin to instead increase cAMP or cytosolic Ca 2+ in hepatocytes. Results: Secretin increased biliary bicarbonate in a dose-dependent fashion and was much more effective when administered via the hepatic artery.Secretin did not affect bile flow.Similarly, ACh increased bicarbonate excretion when infused via the hepatic artery but not the portal vein.The effects of secretin were augmented by ACh, and this was prevented by cyclosporin A.The effects of ACh were blocked by 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS), 5-nitro2-(3-phenylpropylamino)benzoic acid (NPPB), or diphenylamine-2-carboxylic acid (DPC), and the effects of secretin were inhibited by NPPB or DPC and unaffected by DIDS.Neither glucagon nor vasopressin altered biliary bicarbonate. Conclusions: Biliary bicarbonate is regulated by cholangiocytes rather than hepatocytes in normal rat liver.ACh-induced bicarbonate excretion depends on both chloride channels and bicarbonate exchange, whereas secretin-induced bicarbonate excretion is independent of bicarbonate exchange. GASTROENTEROLOGY 2001;121:396-406

Different parameters are considered predictors of bleeding and death in alcoholic cirrhosis. The aim of this study was to establish the prognostic value of a prospective and sequential evaluation of portal pressure, variceal size, and Pugh's score in portal-hypertensive patients with alcoholic cirrhosis but no previous bleeding.Thirty patients were evaluated for 42 +/- 5 months (median, 39 months). After baseline studies, 30 patients underwent an additional evaluation (follow-up 1; median, 10 months), 20 patients a second evaluation (follow-up 2; median, 25 months), and 13 patients a third evaluation (follow-up 3; median, 45 months). No prophylactic treatment for bleeding was given. End points were bleeding and/or death.Seventeen patients died, and 10 patients bled. At follow-up 1, portal pressure decreased both in survivors and nonbleeders (from 18.7 +/- 1.0 to 15.2 +/- 1.3 mm Hg [P0.01] and from 18.9 +/- 0.8 to 16.5 +/- 1.0 mm Hg [P0.05], respectively). On multivariate analysis (Cox model), portal pressure at follow-up 1 had the best prognostic and independent value for both bleeding and survival. Subsequent studies showed similar trends.Measurements of portal pressure provide unique prognostic information for predicting portal hypertensive-related bleeding and mortality in patients with alcoholic cirrhosis.

Hepatic vein thrombosis is thought to be manifested by ascites, abdominal pain, and hepatomegaly, with a uniformly poor prognosis. However, new imaging techniques allow for the diagnosis of hepatic vein thrombosis in asymptomatic cases. The aim of our study was to re-evaluate symptoms and prognosis in patients with hepatic vein thrombosis.Eighty-one patients with hepatic vein thrombosis were analyzed. Forty-seven patients were admitted from 1970 to June 1987 (group I, before Doppler ultrasonography and magnetic resonance imaging were introduced at our hospital) and 34 from July 1987 to June 1991 (group II).When comparing the two groups, age, sex ratio, and causes of hepatic vein thrombosis did not differ. Eight group II patients (asymptomatic patients) had no ascites, hepatomegaly, or abdominal pain. One major hepatic vein remained patent in 41% of group II patients, compared with 12% in group I (P0.05). Intrahepatic collaterals were seen in 79% of group II patients, compared with 21% of group I patients (P0.01). All asymptomatic patients had large intrahepatic and portasystemic collaterals. At 3 years, death occurred in 22% of group II patients and in 45% of group I patients. No asymptomatic patient died.Asymptomatic hepatic vein thrombosis is associated with the spontaneous development of large intrahepatic and portosystemic collaterals. In asymptomatic patients, prognosis at 3 years seems to be good, and surgical therapy may not be required.

PHT 3/10 (33.3%) 4/10 (40%) 1/10 (10%) VT 51/244 (20.9%) 31/244 (12.7%)a 38/244 (15.5%) Controls 45/259 (17.4%) 16/259 (6.3%)b 15/259 (5.8%)

Background & AimsAnimal studies suggest that endocannabinoids could contribute to the development of nonalcoholic fatty liver disease (NAFLD). In addition, NAFLD has been shown to be associated with multiple changes in lipid concentrations in liver biopsies. There are no data on splanchnic free fatty acid (FFA), glycerol, ketone body, endocannabinoid, and lipid fluxes in vivo in subjects with NAFLD.MethodsWe performed hepatic venous catheterization studies in combination with [2H2]palmitate infusion in the fasting state and during a low-dose insulin infusion in 9 subjects with various degrees of hepatic steatosis as determined using liver biopsy. Splanchnic balance of endocannabinoids and individual lipids was determined using ultra performance liquid chromatography coupled to mass spectrometry.ResultsConcentrations of the endocannabinoid 2-arachidonoylglycerol were higher in arterialized (91 ± 33 μg/L basally) than in hepatic venous (51 ± 19 μg/L; P < .05) plasma. Fasting arterial (r = 0.72; P = .031) and hepatic venous (r = 0.70; P = .037) concentrations of 2-arachidonoylglycerol were related positively to liver fat content. Analysis of fluxes of 85 different triglycerides showed that the fatty liver overproduces saturated triglycerides. In the plasma FFA fraction in the basal state, the relative amounts of palmitoleate and linoleate were lower and those of stearate and oleate were higher in the hepatic vein than in the artery. Absolute concentrations of all nontriglyceride lipids were comparable in arterialized venous plasma and the hepatic vein both in the basal and insulin-stimulated states.ConclusionsThe human fatty liver takes up 2-arachidonoylglycerol and overproduces triacylglycerols containing saturated fatty acids, which might reflect increased de novo lipogenesis.

It is generally believed that membranous obstruction of the inferior vena cava in the Budd-Chiari syndrome is caused by congenital malformation. However, it does not explain the late onset of the disease. In the current study, hepatic portion of the inferior vena cava and hepatic veins were studied in 17 autopsy cases of the Budd-Chiari syndrome, 16 of which had no demonstrable cause (idiopathic). A sufficient amount of vena cava tissue was available for evaluation in 15 cases. Nine had membranous obstruction, with thickness varying from 3 to 8 mm. Thrombus formation was recognized in 7 of 9 cases. Occlusion of hepatic vein orifices of varying degree was present in 8 cases. In these occluding lesions, the basic structure of the venous wall was maintained. The intima was transformed into a fibrous laminar structure, and organized thrombi of varying ages were recognized; they were a mixture of fresh thrombi, organized thrombi, fibrous tissues, recanalizations, and calcifications. It is concluded that in these cases of the Budd-Chiari syndrome, occluding and stenosing lesions in the inferior vena cava and hepatic veins were thrombosis and its sequelae. There was no indication of congenital malformation.

Thirteen patients with alcoholic cirrhosis had splanchnic and systemic hemodynamics assessed before and after ingestion of a standard liquid meal of 700 kcal (consisting of isocaloric proteins, lipids, and carbohydrates). Half of the patients (n = 6) were randomized to a treatment group receiving intravenous infusion of propranolol in combination with the meal. No significant effects were observed on systemic hemodynamics after the meal alone. Heart rate (-14%; P less than 0.01) and cardiac index (-24%; P less than 0.01) decreased after meal in combination with propranolol. The mean hepatic venous pressure gradient increased significantly after ingestion of the meal alone with a maximal effect after 30 minutes (+13%; P less than 0.05) and returned to baseline values after 2 hours. Meal in combination with propranolol had no significant effect on the hepatic venous pressure gradient. Hepatic blood flow increased substantially after the meal alone with a maximal effect after 30 minutes (+28%; P less than 0.01), whereas no significant effect was observed after meal in combination with propranolol. Azygos blood flow increased significantly after the meal alone (+36%; P less than 0.05), whereas this effect was abolished in combination with propranolol. In conclusion, ingestion of a peroral mixed meal in cirrhotic patients has, contrary to what is observed in normal controls, no effects on systemic hemodynamics. Substantial changes in splanchnic hemodynamics were observed, and these effects were all abolished when the meal was administered in combination with propranolol.

The effects of long-term oral clonidine treatment on hepatic and systemic hemodynamics and on quantitative liver function tests were investigated in 15 patients with alcoholic cirrhosis. Clonidine was administered at a mean dose of 0.33 +/- 0.1 mg/day (mean +/- SD) for a mean period of 64 +/- 10 days. Oral clonidine induced a significant reduction in the hepatic venous pressure gradient from 18.8 +/- 3.0 mm Hg to 15.9 +/- 3.4 mm Hg (P less than 0.001), which was the result of an increase in the free hepatic venous pressure from 5.1 +/- 4.2 mm Hg to 8.7 +/- 3.8 mm Hg (P less than 0.05). In 10 of the 15 patients (67%), the reduction in the hepatic venous pressure gradient was greater than 10% of baseline values. Hepatic blood flow did not change significantly after clonidine treatment. Additionally, treatment with clonidine decreased mean arterial pressure by 15.5% +/- 6% (P less than 0.001), heart rate by 17.7% +/- 7% (P less than 0.001), and cardiac output by 14.6% +/- 7% (P less than 0.001). However, systemic vascular resistance did not change significantly. There were no adverse effects on liver function, as shown by the nonsignificant changes in galactose-elimination capacity (149 +/- 59 vs. 170 +/- 58 mg/min), hepatic clearance of indocyanine green (0.19 +/- 0.10 vs. 0.17 +/- 0.07 L/min), and hepatic intrinsic clearance of indocyanine green (0.23 +/- 0.14 vs. 0.21 +/- 0.1 L/min) before and after clonidine treatment, respectively. In none of the patients was the drug withdrawn because of side effects, although 12 subjects complained of dry mouths. This study suggests that in patients with alcoholic cirrhosis, long-term oral clonidine administration achieves a reduction in the hepatic venous pressure gradient without adverse effects on hepatic blood flow and liver function.

The goals of treatment of the Budd-Chiari syndrome are relief of portal hypertension, relief of inferior vena cava syndrome, if present, and preservation of hepatic function. This study presents a patient with clinical resolution of the Budd-Chiari syndrome after placement of expandable metallic stents in the inferior vena cava and hepatic veins. A 26-year-old man with severe ascites and lower extremity edema but with relatively preserved hepatic function had a small gradient across a suprahepatic caval web, large gradients across an intrahepatic caval stenosis and the left hepatic vein, and an occluded right hepatic vein. Under angiographic control, web and caval stenosis were balloon-dilated, and modified Gianturco expandable metallic stents were placed in the intrahepatic vena cava. The left hepatic vein was dilated twice and a stent was placed. All gradients were completely eliminated. There were no complications and after 1 year, the stents have fully expanded without migration, edema and ascites have resolved, hepatic function has normalized, and the patient has returned to work. This new technique provides a simple, safe, effective, relatively inexpensive, and potentially long-lasting treatment for selected patients with the Budd-Chiari syndrome.

Vasopressin is often used to treat variceal hemorrhage. However, its efficacy is uncertain, and its portal hemodynamic effects in this setting are unknown. Eleven patients with alcoholic liver disease and bleeding varices were given vasopressin (0.2 U/min for the first hour, then 0.4 U/min for 24 hours). Portal pressure was monitored using an indwelling hepatic vein balloon catheter. Seventeen patients with variceal bleeding who remained stable over 26 hours of initial treatment with crystalloid and blood products served as a comparison group. Vasopressin infusion (0.2 U/min) produced a significant decrease in wedged hepatic venous pressure, hepatic venous pressure gradient (wedged minus free hepatic venous pressure), and heart rate. Increases in the rate of infusion did not produce further decreases in the parameters measured, but the changes were sustained over the course of the infusion. Hemodynamics remained stable in the control group. Portal pressure did not increase when vasopressin was abruptly discontinued in the 3 patients in whom postinfusion measurements were made. Vasopressin retains its portal hypotensive effects in the setting of variceal hemorrhage. Tachyphylaxis does not develop over 26 hours, and a "rebound" increase in portal pressure probably does not occur when the infusion is discontinued.

Portal pressure was monitored by means of an indwelling hepatic vein balloon catheter in patients with alcoholic cirrhosis and bleeding varices to determine the safety and feasibility of the technique and its value in predicting recurrence of bleeding. Forty patients were enrolled. Central venous access could not be achieved in 4 patients (10%). Hepatic vein catheterization was accomplished in the remaining 36 patients. Fourteen patients were either later found to have nonalcoholic liver disease or had already received treatment that excluded them from the protocol. The remaining 22 patients, who were treated with blood and fluid replacement, were monitored for up to 72 hours. Portal pressure was greater than 11 mm Hg in all patients (normal, less than 5 mm Hg) and did not change significantly over the 3 days of study. Portal pressure was significantly higher in the 9 patients who continued to bleed or rebled compared with the 13 patients who remained stable. The lowest pressure associated with continued bleeding or rebleeding was 16 mm Hg. Continuous monitoring of portal pressure in patients with bleeding esophageal varices due to alcoholic cirrhosis is safe and feasible and permits rapid stratification of the risk of continued bleeding or early rebleeding.

Computed tomography and real-time ultrasonography may not be conclusive for the diagnosis of the Budd-Chiari syndrome; in many cases more information may be needed, especially on vascular alterations. Doppler ultrasonography provides qualitative data on flow direction and pattern, thereby contributing significantly to diagnosis. Eight cases in which hepatic vein patency was unclear and presence of intrahepatic vessels resembling hepatic veins raised problems of interpretation in real-time ultrasonography are described. In some cases, patency or occlusion of the upper portion of the inferior vena cava were difficult to identify with real-time ultrasonography. Doppler ultrasonographic investigation showed flow in the hepatic veins to be completely absent in two cases and reversed in another two. In the remaining four cases, a flat waveform was evident. Flow in the inferior vena cava was reversed in four cases and showed a flat waveform in three other cases. Portal vein thrombosis was detected in only one case, whereas the remaining seven patients showed slow hepatopetal flow. These findings demonstrate that absent or reversed flow in the hepatic veins and/or flat flow in the hepatic veins associated with reversed flow in the inferior vena cava may be considered diagnostic for the Budd-Chiari syndrome. For this series the sensitivity of Doppler ultrasonography was 87.5%.

The case of a 33-yr-old woman with an obstruction of small hepatic veins limited to the right lobe of the liver is reported. Abdominal pain, fever, absence of ascites, and normality of liver biochemical tests were the most striking features. Abnormalities characteristic of lesions were observed on surgical wedge liver biopsy, hepatic phlebography, and computerized tomography. None of the usual causes of small hepatic vein obstruction were found. This patient had been taking an oral progestative compound as contraceptive therapy for 7 mo before evaluation. Because progestogens have recently been incriminated in the occurrence of vascular alterations, the possible role of this drug in our case is discussed.

A 31-yr-old man presented with the Budd-Chiari syndrome. A liver biopsy specimen demonstrated noncaseating centrilobular granulomas and an idiopathic necrotizing granulomatous venulitis involving hepatic vein radicles with superimposed thrombotic occlusion. High-dose prednisone therapy produced dramatic clinical improvement with resolution of the hepatomegaly, ascites, and portal hypertension and normalization of the hepatic venous angiogram. A repeat liver biopsy specimen showed resolution of the venulitis and disappearance of the hepatic vein thrombosis. Idiopathic granulomatous venulitis that is reversible with prednisone therapy should be considered in the differential diagnosis of the Budd-Chiari syndrome.

In a review of 906 consecutive liver biopsies, sinusoidal dilatation, unrelated to passive congestion of the liver, sinusoidal infiltration, or cirrhosis, was found in 26 cases (2.9%). In 21 of them the final diagnosis was a neoplastic or granulomatous disease (tuberculosis, brucellosis, Crohn's disease), but in only half of them was there evidence of neoplastic or granulomatous infiltration of the liver. In the remaining cases, sinusoidal dilatation was either the only histological abnormality or it was associated with nonspecific changes. Although the pathogenesis of sinusoidal ectasia is not known, our findings indicate a definite relationship to the presence of tumor or granulomatous disease in the liver or elsewhere in the body. It is concluded that the finding of sinusoidal dilatation as an isolated histological change in a liver biopsy specimen should prompt the search for a tumor or a disease associated with granulomas.

Twenty cirrhotic patients with esophageal varices and an episode of gastrointestinal bleeding within the previous week underwent hemodynamic studies which used an indicator dilution technique based on selective catheterization of the splanchnic arteries and hepatic vein. Portasystemic shunting of splenic arterial flow averaged 95% (range 80 to 100%) in 17 patients with proved variceal bleeding, but only 8% in 3 who bled from another site. Superior mesenteric shunting averaged 70% (range 27 to 100%) in those who bled from varices, and no shunting was detectable in the nonvariceal bleeders. In at least 4 patients shunting was essentially complete from both beds. In 11 others, however, more than one-third of mesenteric flow still perfused the liver. The pressure gradient from the wedge to free position in the hepatic vein and the hepatic blood flow bore no relationship to the degree of shunting. These data indicated the variability of the splanchnic hemodynamic pattern that may exist in alcoholic patients with esophageal varices. If hemodynamic factors are important in determining the response to surgery, indicator dilution studies to quantitate portasystemic shunting may be a vital part of preoperative evaluation.

Systemic and splanchnic hemodynamics, renal blood flow, and renal function were studied in 13 patients with cirrhosis both before and 1 h after oral administration of 40 mg of propranolol (acute administration) and 1 mo after continuous administration of this substance at doses reducing the heart rate by 25% (chronic administration). Cardiac output and the gradient between wedged and free hepatic venous pressures significantly decreased after acute and chronic administration of propranolol; mean arterial pressure did not change significantly and systemic vascular resistance significantly increased. Renal blood flow and renal vascular resistance did not change significantly after acute administration of propranolol and renal function did not change significantly after acute or chronic administration of propranolol. We conclude that propranolol does not alter renal function in patients with cirrhosis who are in good physical condition.

In patients with alcoholic cirrhosis, wedged hepatic venous pressure closely reflects portal venous pressure. This study was carried out to determine if propranolol-induced reductions in portal venous pressure are accurately evaluated by the measurement of wedged hepatic venous pressure. Hepatic venous cannulation and percutaneous transhepatic catheterization of the portal vein were simultaneously performed in 7 patients with alcoholic cirrhosis. One hour after oral administration of 40 mg of propranolol, wedged hepatic and portal venous pressures significantly decreased from 24.3 +/- 3.5 (mean +/- SD) to 19.0 +/- 3.0 mmHg, and from 24.7 +/- 3.9 to 22.4 +/- 3.6 mmHg, respectively. Although no significant difference was found between baseline wedged hepatic and portal venous pressures, a significant difference was found between these pressures after propranolol administration. We concluded that during acute administration of a drug acting on the splanchnic circulation, the measurement of wedged hepatic venous pressure may not provide a reliable estimation of the magnitude of the changes in portal venous pressure. There is, however, no evidence that the direction of the changes might not be adequately assessed by wedged hepatic venous pressure measurement.

To estimate vascular changes in chronic liver disease, we quantitated intrahepatic arteriovenous and portal-systemic shunts in 12 patients with cirrhosis and arteriovenous shunts alone in 4 patients with cirrhosis. An index was obtained for intrahepatic arteriovenous shunts by instilling technetium 99m-macroaggregated albumin into the proper hepatic artery and portal-systemic shunts, by the same procedure done in the portal trunk, near the porta hepatis on different days. Counts were taken over the liver and both lungs in the anterior as well as the posterior view for calculation of the shunt index: cpm in lungs divided by cpm in liver and lungs X 100%. In the 12 patients with cirrhosis in whom both shunts were measured, intrahepatic arteriovenous shunting was significantly lower compared with intrahepatic portal-systemic shunting (1.4% +/- 1.1% vs. 36.0% +/- 29.0%, p less than 0.001). Thus, it seems that in patients with cirrhosis, the development of intrahepatic arteriovenous shunts is not as great as that of portal-systemic shunts, which were found in this study to be considerable and variable in degree.

Combined hepatic and umbilicoportal catheterization was performed in 43 compensated cirrhotic patients. The umbilicoportal catheter was positioned in the portal vein (41 patients) or the left branch of the portal vein (2 patients). Wedged hepatic venous pressure was equal to free portal venous pressure in 17 patients. The difference between these two parameters was between 0.5 and 1 mm Hg (17 cases). In the other 9 patients, the difference ranged from 1.5 to 4 mm Hg. The correlation between these two parameters was highly significant. These studies show that in patients with cirrhosis, there is a highly significant correlation between wedged hepatic venous pressure and free portal venous pressure.

Gastrinomas were localized by concurrent blood sampling in the hepatic vein and portal vein tributaries in 10 of 12 Zollinger-Ellison patients. In one patient, the portal vein could not be catheterized; in the other, a gastrin gradient could not be picked up in either of two examinations. Six of the patients were subsequently operated upon: 5 had pancreatoduodenal resection and 1 had laparotomy at which metastases were found. Four of the resections were probably curative as the patients have done well without treatment since with concentrations of gastrin in serum near zero. The observation period ranges from 17 to 20 mo. It is concluded that transhepatic catheterization of the hepatic vein and portal vein tributaries with blood sampling for gastrin determination permit the verification of the Zollinger-Ellison diagnosis and the localization of the gastrinomas. The latter may lead to the selection of a curative operative treatment in some of the patients.

Isosorbide dinitrate, a long-acting organic nitrate, has been shown to decrease portal pressure in the experimental animal and humans. We conducted a double-blind randomized hemodynamic evaluation of the effects of placebo and 10 mg and 20 mg isosorbide dinitrate in stable individuals with alcoholic cirrhosis. Baseline values for all three groups were similar. Isosorbide dinitrate resulted in a peak reduction of the hepatic venous gradient of 24.7% +/- 3.0%, with significantly decreased values 4 h after the administration of the 20-mg dose. A reduction of arterial pressure and cardiac index (peak decrease of 25.7% +/- 1.5%) was well tolerated by 13 of 15 patients. Changes in mean arterial pressure were not predictive of modifications in the hepatic vein wedge pressure. There was no relation between the area under the plasma isosorbide dinitrate concentration curve and hemodynamic changes. Levels of isosorbide-5-mononitrate, a vasoactive metabolite, were detectable for an 8-h period. Isosorbide dinitrate significantly reduced portal pressure in stable cirrhotics, in association with systemic hemodynamic changes. Thus, titration of isosorbide dinitrate is required to maximize hemodynamic benefits in individual patients. As the decrease in portal pressure is more predictable than the effect of previously tested pharmacologic agents, isosorbide dinitrate should be evaluated for its efficacy in the management of portal hypertension.

The accuracy and reliability of a balloon catheter for measuring wedged hepatic venous pressure (WHVP) were evaluated in 82 simultaneous determinations using the balloon catheter technique and the direct measurement of portal venous pressure. These measurements showed a close positive correlation (r = 0.97) over a wide range of pressures in 12 normal and 4 cirrhotic dogs. Studies were then performed in 13 cirrhotic patients in whom the balloon catheter was introduced using the Seldinger technique. Free hepatic venous pressure (FHVP) was measured with the balloon undistended. By distending the balloon to occlude the hepatic vein, the WHVP was recorded. Comparison of FHVP and WHVP with the conventional and balloon techniques revealed a close positive correlation (r = 0.89 and 0.93, respectively). The correlation was virtually perfect for the hepatic venous pressure gradient (WHVP-FHVP, r = 0.98). The balloon technique offers many advantages over the conventional approach, including the ease of achieving and demonstrating the wedged position from the femoral approach, the ability to measure the free and wedged pressure repeatedly without manipulating the catheter, and the measurement of pressure in a larger, more representative segment of the liver.

In 6 anesthetized dogs receiving 400 pmoles kg-1 hr-1 of natural human little gastrin intravenously, blood samples for radioimmunoassay of gastrin were taken repeatedly from a carotid artery, jugular vein, femoral vein, renal vein, and mesenteric vein. The calculated removal of gastrin in a single passage through these four vascular beds ranged from 21 to 30% and did not differ significantly among them. In 6 additional dogs, the carotid arterial to hepatic venous removal was 40%. We conclude that the percentage removal of gastrin is about the same in all of the major vascular beds of the body.

We report 3 cases of hepatic venocclusive disease occurring in renal transplant patients receiving azathioprine and combine our experience with 4 other previously reported cases. The data suggest a clinical syndrome characterized by (a) delayed clinical onset, (b) striking male predominance, (c) presentation with jaundice followed by evidence of portal hypertension, and (d) poor prognosis. One of our patients, who is still alive 40 mo after the first onset of symptoms of liver disease, showed striking clinical improvement with discontinuation of azathioprine and subsequent deterioration on reinstitution. We suggest that azathioprine may be closely linked with the development of venocclusive disease in renal transplant patients and that the frequency of this disorder may be more common than previously reported. To attempt to prevent a fatal outcome, this group of patients should be closely monitored for the earliest signs of hepatic venocclusive disease through periodic serum bilirubin and alkaline phosphatase determinations. Patients with abnormal tests should undergo liver biopsy. If hepatic venocclusive disease is found, prompt withdrawal of azathioprine is indicated.

The nature and significance of vascular lesions in alcoholic liver disease were studied in 200 autopsies. Three principal types of lesions were recognized: (a) Lymphocytic phlebitis, consisting of a chronic inflammatory cell infiltrate of the wall of terminal hepatic venules (central veins) or intercalated (sublobular) veins, was noted in 16.7% of patients with precirrhotic alcoholic hepatitis and 4.3% of patients with cirrhosis. (b) Phlebosclerosis, consisting of perivenular scarring with gradual obliteration of the lumen of terminal hepatic venules and sometimes intercalated veins was found to some degree in all patients with alcoholic hepatitis or cirrhosis. (c) Veno-occlusive lesions, consisting of intimal proliferation, fibrosis, and narrowing of the lumen of terminal hepatic venules, intercalated veins, and occasionally portal veins were found in 52.1% of cases of precirrhotic alcoholic hepatitis with total occlusion of some terminal hepatic venules or intercalated veins, or both, in 14.6%. In alcoholic cirrhosis, veno-occlusive lesions were present to some degree in 74.1% with totally occluded vessels found in 46.8%. Evidence of portal hypertension was present in 47.9% of patients with precirrhotic alcoholic hepatitis and was significantly associated with the degree of both veno-occlusive change and phlebosclerosis, which tend to occur together. It is concluded that both veno-occlusive lesions and phlebosclerosis contribute to the development of portal hypertension in alcoholic liver disease. Veno-occlusive lesions in the cirrhotic liver may contribute to atrophy, with loss of functioning parenchyma. The etiopathogenesis of the vascular lesions in alcoholic liver disease requires further investigation.