Effects of Long-term Propranolol and Octreotide on Postprandial Hemodynamics in Cirrhosis: A Randomized, Controlled Trial

Transcription

1 GASTROENTEROLOGY 2002;122: Effects of Long-term Propranolol and Octreotide on Postprandial Hemodynamics in Cirrhosis: A Randomized, Controlled Trial JULIO D. VOROBIOFF,* MARCELO GAMEN,* DAVID KRAVETZ, EDUARDO PICABEA,* ROBERTO VILLAVICENCIO,* JUAN BORDATO,* ANDRÉS RUF,* FERNANDO BESSONE,* GUSTAVO ROMERO, JORGE PALAZZI, ALICIA NICORA, MARÍA PASSAMONTI,* and HUGO TANNO* *Liver Unit and Hepatic Hemodynamic Laboratory, Fundación Dr. J. R. Villavicencio & Sanatorio Parque, Rosario; Hospital Bonorino Udaondo, Buenos Aires; and the Instituto de Estudios Bioquímicos, Rosario, Argentina Background & Aims: Postprandial increases in portal pressure may influence esophageal variceal rupture. The effects of chronic propranolol and octreotide (100 and 200 g subcutaneously in a single dose) on postprandial hemodynamics were evaluated. Methods: 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 g), or (3) octreotide (200 g) (n 12 for each group). Thirty minutes postinjection a new baseline was established and measurements were repeated 30 and 60 minutes after the meal. Results: First study: Baseline portal pressure was mm Hg, 30 and 60 minutes after the meal it was mm Hg and mm Hg, respectively (both P < 0.01 vs. baseline). Cardiac index (CI) was , , and L min 1 m 2, respectively (both P < 0.05 vs. baseline). Peripheral vascular resistance was , (P NS), and dynes sec cm 5 (P < 0.05 vs. baseline), respectively. Second study: Propranolol and placebo did not blunt postprandial increase in portal pressure. Octreotide (100 g) partially ameliorated postprandial increase in portal pressure. Octreotide (200 g) significantly enhanced the portal hypotensive effect of propranolol and blunted the postprandial increase in portal pressure. Conclusions: Octreotide blunts postprandial increase in portal pressure not prevented by long-term propranolol administration. An increase in portal pressure has long been recognized in cirrhotic patients during the postprandial period. 1 6 This is presumably caused by an augmentation in splanchnic blood flow, a phenomenon known as postprandial hyperemia, 7 9 and is maximally observed minutes after the ingestion of food. Several vasodilatory peptides, including glucagon and vasoactive intestinal peptide, reach peak levels after meals and are therefore thought to play a major role in postprandial hyperemia. 8,10 These sudden increases or peaks in portal pressure, in an already hypertensive territory, could be involved in esophageal variceal rupture and bleeding by increasing variceal wall tension. 11 Propranolol, a nonselective adrenergic blocker, reduces portal pressure by decreasing portal blood flow. 12,13 Its effects on portal pressure fluctuations have been reported. 14,15 Specifically, during the postprandial period, acute administration of propranolol did not blunt postprandial hyperemia 16 but, in another study, it inhibited postprandial increases in hepatic blood flow and portal pressure. 17 A third study showed that long-term propranolol administration does not prevent postprandial increases in portal pressure in cirrhotic patients. 18 So, available information regarding the effects of propranolol on postprandial portal hemodynamics is scanty and controversial. Octreotide, a long-acting synthetic somatostatin analogue, 19 was introduced for clinical use in patients with cirrhosis and variceal hemorrhage. 20 Its beneficial effect in the control of acute variceal hemorrhage seems to be owing to splanchnic vasoconstriction and, thus, decreased portal blood flow. 21 Additionally, and perhaps more important, is the effect of octreotide in blunting the postprandial hyperemic response in cirrhotic patients This could be mediated by its known inhibitory action on vasodilatory peptide release during the immediate postprandial period. 10,23,25 Abbreviations used in this paper: Bas 1, baseline hemodynamic data; Bas 2, repeated measure of hemodynamic data; CI, cardiac index, CO, cardiac output; HR, heart rate; HVPG, hepatic venous pressure gradient; MAP, mean arterial pressure; PVR, peripheral vascular resistance by the American Gastroenterological Association /02/$35.00 doi: /gast

2 April 2002 POSTPRANDIAL HEMODYNAMICS IN CIRRHOTIC PATIENTS 917 Table 1. Baseline Clinical Characteristics of Studied Patients Patients Age (yr) Sex (men/women) Etiology (ethyl/other) Previous bleeding (no/yes) Variceal size (1 3) Pugh s score HVPG (mm Hg) Propranolol and placebo (n 12) /3 7/5 7/ Propranolol and octreotide 100 g (n 12) /2 9/3 6/ Propranolol and octreotide 200 g (n 12) /5 4/8 7/ Patients not included (n 10) /2 4/8 7/ a NOTE. Data are shown as mean SEM. n 46. a P 0.05 vs. Propranolol and octreotide 200 g. Therefore, this study was designed to: (1) evaluate the effects of chronic administration of propranolol on splanchnic and systemic hemodynamics in the postprandial period, and (2) compare these effects with those observed after the addition of octreotide to chronic adrenergic blockade, in a group of portal hypertensive cirrhotic patients. Materials and Methods Patients Forty-six patients with cirrhosis, portal hypertension, and esophageal varices were included in the study. Thirty-four patients were men and 12 were women; the mean age was 52 3 years (mean SEM). Nineteen patients had previously bled from varices; whereas the remaining 27 had not. The exclusion criteria were: age less than 18 or greater than 80 years, previous pharmacologic or endoscopic treatment for portal hypertension, presence of hepatocellular carcinoma, contraindications to -blockers, and refusal to participate in the study. The severity of liver disease was graded according to Pugh s criteria. 26 Baseline clinical data of the 46 patients studied are shown in Table 1. The present investigation was approved by the Human Research Committee of Fundación Dr. J. R. Villavicencio and patients gave informed written consent to participate in the study. Study Design First study: baseline hemodynamic evaluation. Portal pressure was determined by the hepatic venous pressure gradient (HVPG), which is obtained by subtracting the free hepatic venous pressure from the wedged (occluded) hepatic venous pressure (Figure 1). After an overnight fast, right hepatic vein catheterization was performed percutaneously through the femoral vein, and pressure was recorded in both the wedged (occluded) and the free position by using a balloon catheter (Medi Tech, Cooper Scientific Corp, Watertown, MA). 27 The wedged position was checked by the absence of reflux after the injection of 2 ml of contrast medium. Pressures were measured by using a strain gauge transducer previously calibrated and recorded on paper. Each pressure reading was recorded at least twice, and an electronic mean was obtained. The pulmonary artery was catheterized through the femoral vein with a Swan-Ganz catheter (Edwards Laboratory, Los Angeles, CA) and cardiac output (CO) was measured by thermodilution. Each measurement was performed in triplicate, and an average was taken. Cardiac index (CI) was calculated by dividing the CO by the patient s body surface. Arterial pressure (with an external sphygmomanometer) and heart rate (HR) were also determined. Mean arterial pressure (MAP) was calculated according to the following formula: MAP systolic diastolic 2 /3 Peripheral vascular resistance (PVR) was calculated according to the following formula: PVR dynes.sec.cm 5 MAP RAP /CO , where RAP indicates right atrial pressure. After baseline hemodynamic parameters were obtained, a standard mixed liquid meal was given. The test meal consisted of 237 ml of a balanced dietary supplement (Ensure Plus, Abbot Laboratories, North Chicago, IL), which supplied 355 kcal as 14.7 g protein, 53.3 g carbohydrate, and 32 g lipid. The meal was consumed over a period of 10 minutes. Hemo- Figure 1. Study design.

3 918 VOROBIOFF ET AL. GASTROENTEROLOGY Vol. 122, No. 4 dynamic measurements were repeated 30 and 60 minutes later. After the study was completed, all patients were started on oral propranolol therapy. Propranolol was initially given at a dose of 20 mg twice a day and this dose was increased step-wise at 4 6-day intervals until the heart rate was reduced by 25% or to 55 beats/minute. Second study: response to chronic administration of propranolol plus octreotide. Thirty-six of the 46 patients entered into the study underwent repeat hemodynamic measurements after 39 2 days of treatment. Ten patients did not undergo repeat measurements: 1 required a liver transplantation, 2 did not respond to the standard meal in the initial study (the HVPG did not change at 30 and 60 minutes), 3 had variceal bleeding, 2 were lost to follow-up, and 2 others were noncompliant with treatment. The mean dose of propranolol received by the 36 patients was mg/d (range, ). Measurements were performed after a minimum of 4 hours after propranolol administration. Once baseline hemodynamic data were obtained (Bas 1 ), patients were randomly assigned to 1 of the 3 following groups (n 12 each group): (1) placebo (1 ml distilled water), (2) octreotide 100 g, or (3) octreotide 200 g (Sandostatin; Novartis, Argentina) subcutaneously. Patients were randomized by means of sealed and consecutively numbered envelopes containing the allocated therapy according to a randomization code generated by computer. Randomization and administration of the assigned treatment was performed by a third person. All other participants were and remained blind until study completion. Thirty minutes after injections, measurements were repeated (Bas 2 ). Then, the standard meal was given and all measurements were repeated 30 and 60 minutes later. Determination of octreotide level. Blood samples for octreotide serum concentrations were processed by Novartis Pharmaceuticals (East Hanover, NJ), as previously described. 28 Statistical analysis. Data are shown as mean SEM. Differences between means were analyzed by paired or unpaired Student t test or analysis of variance as appropriate. Results were considered significant at P Results First Study: Baseline Hemodynamic Evaluation Hepatic vein catheterization. Mean baseline HVPG in the 46 patients included in the study was mm Hg. It increased to mm Hg and to mm Hg at 30 and 60 minutes after the standard meal, respectively (both P 0.01 vs. baseline) (mean respective increases of 18.5% 2% and 14.5% 2%). In patients with previous variceal and/or portal hypertensive gastropathy bleeding (n 19), the mean percentage increase in HVPG at 30 minutes after the meal was higher than that observed in previously nonbleeders (n 27) (22.1% 3% vs. 15.9% 2%, respectively) (P 0.06). At 60 minutes there were no differences between both groups (13.7% 3% vs. 15.0% 3%). A higher increase in HVPG at 30 minutes was also observed in patients with esophageal varices grade 2 3 (n 36) when compared with those with esophageal varices grade 1 (n 10): 20.0% 2% vs. 13.1% 3% (P 0.06). At 60 minutes there were no differences between both groups (15.8% 3% vs. 9.7% 3%). Alcoholic (n 24) and nonalcoholic patients (n 22) had similar changes in HVPG (19.2% 2% vs. 17.7% 2% and 13.7% 3% vs. 15.4% 3% at 30 and 60 minutes, respectively), as did patients categorized as Child Pugh grade A (n 19) and those grade B C(n 27) (19.5% 2% vs. 17.8% 2% and 14.0% 3% vs. 14.8% 3% at 30 and 60 minutes, respectively). No significant differences were observed between men (average weight, 76 2 kg) and women (average weight, 63 3 kg). Mean baseline HVPG in the 36 patients that completed the study was mm Hg. It increased to mm Hg and to mm Hg (both P 0.01 vs. baseline) (mean respective increases of 19.3% 2% and 13.9% 2%) at 30 and 60 minutes after the meal, respectively. Systemic hemodynamics. Mean baseline CI in the 46 patients included in the study was L min 1 m 2. It increased to and to L min 1 m 2 (both P 0.02 vs. baseline) at 30 and 60 minutes after the meal, respectively. Baseline PVR was dynes sec cm 5. It decreased to dynes sec cm 5 (P NS) at 30 minutes and to dynes sec cm 5 (P 0.05 vs. baseline) at 60 minutes after the meal. Baseline heart rate was 82 2 beats/minute and increased to 84 2 beats/minute (P NS) at 30 minutes and to 85 2 beats/minute (P 0.05 vs. baseline) at 60 minutes after the meal. Baseline MAP was 94 2 mm Hg and decreased to 92 2mmHg(P NS) at 30 minutes and to 91 2mmHg(P NS) at 60 minutes after the meal. In the 36 patients that completed the study, mean baseline CI was L min 1 m 2. It increased to L min 1 m 2 and to L min 1 m 2 (both P 0.05 vs. baseline) at 30 and 60 minutes after the meal, respectively (Figure 2). Baseline PVR was dynes sec cm 5. It decreased to dynes sec cm 5 (P NS) at 30 minutes and to dynes sec cm 5 (P 0.05 vs. baseline) at 60 minutes after the meal (Figure 2). Baseline HR was 84 2 beats/minute and increased to 85 2 beats/ minute (P NS) at 30 minutes and to 86 2 beats/ minute (P 0.05 vs. baseline) at 60 minutes after the

4 April 2002 POSTPRANDIAL HEMODYNAMICS IN CIRRHOTIC PATIENTS 919 Figure 2. Baseline hemodynamic evaluation. CI and PVR in the 36 patients that completed the study. *P 0.05 vs. 0 minutes. Figure 4. Propranolol and octreotide 100 g(n 12 patients). HVPG at baseline and at the second hemodynamic evaluation. Baseline study: 30 and 60 minutes after the meal HVPG increased significantly (*P 0.01 vs. 0 minutes). Second study: After octreotide 100 g, HVPG increased significantly at 60 minutes after the meal ( P 0.02 vs. Bas 2 ). **P 0.02 vs. 0 minutes at baseline study. meal. Baseline MAP was 97 2 mm Hg and decreased to 95 2mmHg(P NS) at 30 minutes and to 93 2mmHg(P NS) at 60 minutes after the meal. Second Study Mean propranolol dose and treatment time period in the 3 groups of patients were: propranolol and placebo ( mg/d; 36 3 days), propranolol and octreotide 100 g ( mg/d; 39 3 days), and propranolol and octreotide 200 g ( mg/d; 45 4 days) (P NS). Propranolol and Placebo (n 12) Hepatic vein catheterization. Mean Bas 1 (preplacebo) HVPG was mm Hg. Bas 2 (postplacebo) was mm Hg (P NS). Thirty and 60 minutes after the meal it was mm Hg and mm Hg (both P 0.01 vs. Bas 2 ) (mean respective increases of 19.4% 6% and 20.7% 6%) (Figure 3). Figure 3. Propranolol and placebo (n 12 patients). HVPG at baseline and at the second hemodynamic evaluation. Baseline study: 30 and 60 minutes after the meal HVPG increased significantly (*P 0.01 vs. 0 minutes). Second study: After placebo, HVPG increased significantly at 30 and at 60 minutes after the meal ( P 0.01 vs. Bas 2 ). **P 0.05 vs. 0 minutes at baseline study. Systemic hemodynamics. Mean Bas 1 (preplacebo) CI was L min 1 m 2. Bas 2 (postplacebo) was L min 1 m 2 (P NS). Thirty and 60 minutes after the meal it was L min 1 m 2 (P NS) and L min 1 m 2 (P NS), respectively. Mean Bas 1 (preplacebo) PVR was dynes sec cm 5. Bas 2 (postplacebo) was dynes sec cm 5 (P NS). Thirty and 60 minutes after the meal it was dynes sec cm 5 (P NS) and dynes sec cm 5 (P NS), respectively. Mean Bas 1 (preplacebo) MAP was 94 3 mm Hg. Bas 2 (postplacebo) was 93 4mmHg (P NS). Thirty and 60 minutes after the meal it was 92 4mmHg(P NS) and 91 3mmHg(P NS), respectively. Mean Bas 1 (preplacebo) HR was 60 2 beats/minute. Bas 2 (postplacebo) was 61 1 beats/ minute (P NS). Thirty and 60 minutes after the standard meal it was 64 2 beats/minute (P 0.01 vs. Bas 2 ) and 63 2 beats/minute (P NS), respectively. Propranolol and Octreotide 100 g (n 12) Hepatic vein catheterization. Mean Bas 1 (preoctreotide) HVPG was mm Hg. Bas 2 (postoctreotide) was mm Hg (P NS). Thirty and 60 minutes after the meal it was mm Hg (P NS) and mmHg(P 0.02 vs. Bas 2 )(a mean increase of 13.4% 5%), respectively (Figure 4). Systemic hemodynamics. Mean Bas 1 (preoctreotide) CI was L min 1 m 2. Bas 2 (postoctreotide) was L min 1 m 2 (P NS). Thirty and 60 minutes after the meal it was (P NS) and L min 1 m 2 (P NS), respectively. Mean Bas 1 (preoctreotide) PVR was dynes sec cm 5. Bas 2 (postoctreotide) was dynes sec cm 5 (P NS). Thirty and 60 minutes

5 920 VOROBIOFF ET AL. GASTROENTEROLOGY Vol. 122, No. 4 Figure 5. Propranolol and octreotide 200 g(n 12 patients). HVPG at baseline and at the second hemodynamic evaluation. Baseline study: 30 minutes after the meal HVPG increased significantly (*P 0.01 vs. 0 minutes). Second study: After octreotide 200 g, HVPG decreased significantly ( P 0.05 vs. Bas 1 ). A moderate but significant increase is seen at 60 minutes after the meal ( P 0.05 vs. Bas 2 ). Note that HVPG value at 60 minutes is similar to that observed at Bas 1.**P 0.01 vs. 0 minutes at baseline study. after the meal it was dynes sec cm 5 (P NS) and dynes sec cm 5 (P NS), respectively. Mean Bas 1 (preoctreotide) MAP was 88 3 mm Hg. Bas 2 (postoctreotide) was 90 3mmHg(P NS). Thirty and 60 minutes after the meal it was 88 3mmHg(P NS) and 89 3mmHg(P NS), respectively. Mean Bas 1 (preoctreotide) HR was 59 2 beats/minute. Bas 2 (postoctreotide) was 62 2 beats/ minute (P 0.05 vs. Bas 1 ). Thirty and 60 minutes after the standard meal it was 64 3 beats/minute (P NS) and 65 2 beats/minute (P 0.05 vs. Bas 2 ), respectively. Propranolol and Octreotide 200 g (n 12) Hepatic vein catheterization. Mean Bas 1 (preoctreotide) HVPG was mm Hg. Mean Bas 2 (postoctreotide) was mm Hg (P 0.05 vs. Bas 1 ) (a mean decrease of 7.8% 4%). Thirty and 60 minutes after the meal it was mm Hg (P NS) and mm Hg (P 0.05 vs. Bas 2 )(amean increase of 10% 3%), respectively (Figure 5). Systemic hemodynamics. Mean Bas 1 (preoctreotide) CI was L min 1 m 2. Bas 2 (postoctreotide) was L min 1 m 2 (P NS). Thirty and 60 minutes after the meal it was L min 1 m 2 (P NS) and L min 1 m 2 (P NS), respectively. Mean Bas 1 (preoctreotide) PVR was dynes sec cm 5. Bas 2 (postoctreotide) was dynes sec cm 5 (P NS). Thirty and 60 minutes after the meal it was dynes sec cm 5 (P NS) and dynes sec cm 5 (P NS), respectively. Mean Bas 1 (preoctreotide) MAP was 99 4 mm Hg. Bas 2 (postoctreotide) was 96 4mmHg(P NS). Thirty and 60 minutes after the meal it was 96 4mmHg(P NS) and 96 4 mm Hg (P NS), respectively. Mean Bas 1 (preoctreotide) HR was 61 1 beats/minute. Bas 2 (postoctreotide) was 62 1 beats/minute (P NS). Thirty and 60 minutes after the meal it was 62 2 beats/minute (P NS) and 62 1(P NS) beats/minute, respectively. Serum octreotide levels. The mean plasma concentrations of octreotide at 30, 60, and 90 minutes were , , and pg/ml, and , , and pg/ml after the administration of 100 and 200 g, respectively (P NS). Discussion The HVPG behavior observed in these patients reveals that the postprandial increase in portal pressure is an almost universal phenomenon in portal hypertensive cirrhotic patients. No increase was seen in only 2 of 46 patients after the standard meal administration. This response is mainly caused by an increase in the wedged hepatic venous pressure and has a similar profile in each patient, reaching its peak 30 minutes after the meal and declining slightly thereafter, but still remaining significantly higher than baseline after 60 minutes. These results are quantitatively similar to those previously reported in cirrhotic patients. 2 6,17,21,24 Some minor discrepancies, regarding differences in portal pressure percentage increases and the precise postprandial pressure peak time, are probably related to the different volume and composition of administered meals and to diverse experimental designs adopted in different studies. The HVPG response pattern was homogeneous irrespective of liver disease etiology and/or patient s Child-Pugh status. An interesting and provocative finding is the higher HVPG postprandial increase observed in patients with previous portal hypertensive related bleeding as compared with nonbleeders and a similar trend shown by patients with grade 2 3 esophageal varices when compared with those with small-size varices. If postprandial HVPG peaks play a role in determining variceal bleeding, this finding would in part support the observation of a higher rebleeding or bleeding risk in prior bleeders or those patients with large varices, respectively. 12,29 Systemic hemodynamic changes during the first study were also remarkable. CO increased during the postprandial period, these increases were significant 30 and 60 minutes after the meal. Small, nonsignificant decreases in MAP accompanied CO variations. Consequently, PVR showed a progressive reduction that became significant 1 hour after the meal, as previously shown in a similar study. 24 These systemic events could be caused by the same humoral factors mediating splanchnic vasodilation

6 April 2002 POSTPRANDIAL HEMODYNAMICS IN CIRRHOTIC PATIENTS 921 because they occur synchronously with these changes. Glucagon has been recognized as one of the main splanchnic vasodilatory peptides released in the postprandial period. 4,22,24,30,31 However, systemic hemodynamic effects of glucagon in cirrhotic patients, though significant, are mild. 32 Other peptides, simultaneously released with glucagon, or substances such as endothelial-derived relaxing factors (i.e., nitric oxide), may also be mediators of systemic vasodilatory effects. 30,31 Long-term propranolol administration did not prevent the postprandial increase in portal pressure, patients having a mean HVPG increase similar to that observed during the baseline study. Moreover, this increase was similar in propranolol responders (n 8) and nonresponders (n 4) (data not shown). These results indicate that -adrenergic blockade does not interfere with the basic mechanism that induces the postprandial hyperemic state. Propranolol decreases portal venous inflow through -1 and -2 adrenergic blockade without modifying mesenteric vasodilatation caused by food ingestion, which is mainly mediated by hormonal mechanisms. 33,34 Our results are consistent with previous studies showing that propranolol does not blunt postprandial hyperemia, 16 nor does it prevent the postprandial increase in portal pressure in cirrhotic patients. 18 However, because of the lower fasting HVPG level induced by propranolol, the absolute value of postprandial HVPG was significantly lower than that observed in the baseline study, despite a similar percentage increase (Figure 3). This may constitute a protective effect of propranolol in portal hypertensive patients. Octreotide administration modified postprandial HVPG response in patients on long-term propranolol. The 100- g dose induced a delay in postprandial HVPG peak, being this significant only at 60 minutes after the meal. In patients receiving the dose of 200 g, 2 relevant hemodynamic features were observed. First, there was an enhancement of the portal hypotensive effect of propranolol, thereby allowing a baseline HVPG (Bas 2 ) significantly lower than that given by propranolol alone (Bas 1 ). This finding could be related to a recent experimental report, showing that octreotide potentiates local vasoconstriction in isolated, preconstricted superior mesenteric arteries. 35 Nonselective adrenergic blockers, through the blockade of 2 -receptors, induce splanchnic and peripheral vasoconstriction by allowing unopposed effects of catecholamines on -adrenoreceptors. 36 Then, splanchnic arterial vessels of patients receiving propranolol may be preconstricted, thereby augmenting octreotide vasoactive effects. However, a direct effect of octreotide cannot be excluded, given the results shown by other investigators after intravenous administration. 37 Second, though HVPG level at 60 minutes was modestly but significantly higher than Bas 2, it did not exceed the initial Bas 1 value. Therefore, the net result is a neutralization of postprandial increase in portal pressure (Figure 5). This last effect is probably caused by octreotide inhibitory effect on the release of vasodilatory peptides mediating postprandial hyperemia. 10,23,25 Differences in splanchnic hemodynamic effects among both groups of patients may be related to serum octreotide concentrations. These were higher in patients receiving 200 g and were maintained for at least 90 minutes. The same dose, reaching similar serum octreotide levels, was reported to blunt postprandial splanchnic hyperemia in cirrhotic patients. 22 The systemic hemodynamic response after the meal was significantly decreased by propranolol. Although both increases in CO and reduction in peripheral vascular resistance were observed, these change were mild and not significant. Based on the lack of effect of propranolol in preventing postprandial HVPG peaks and its simultaneous amelioration of systemic circulatory changes, one could postulate that postprandial splanchnic and systemic hemodynamic phenomena are mediated by different mechanisms. Splanchnic hyperemia, being mediated by the release of vasodilatory peptides, is not influenced by propranolol. The systemic response seen in nontreated patients could be an adaptative response to splanchnic blood pooling, as observed in normal subjects during the postprandial period, 38,39 and this mechanism could be inhibited by -blockers. A local arterial constrictive effect, unrelated to systemic hormonal modulation of glucagon, has been shown for octreotide in cirrhotic patients. 28 However, octreotide plus propranolol made no difference in the systemic response. In conclusion, though decreasing the expression of the hyperdynamic syndrome and markedly reducing HVPG, long-term propranolol administration does not prevent postprandial increases in portal pressure. The addition of octreotide (200 g) enhances the portal hypotensive effect of propranolol and neutralizes postprandial HVPG augmentation. References 1. Castleman L, Brandt JL, Ruskin HD. The effects of oral feeding of meat and glucose on hepatic vein wedge pressure in normal and cirrhotic subjects. J Lab Clin Med 1958;51: Lee SS, Hadengue A, Moreau R, Sayegh R, Hillon P, Lebrec D. Postprandial hemodynamic responses in patients with cirrhosis. Hepatology 1988;8: Gaiani S, Bolondi L, Li Bassi S, Santi V, Zironi G, Barbara L. Effect of meal on portal hemodynamics in healthy humans and in patients with chronic liver disease. Hepatology 1989;9: McCormick PA, Dick R, Graffeo M, Wagstaff D, Madden A, McIntyre N, Burroughs AK. The effect of non-protein liquid meals on

7 922 VOROBIOFF ET AL. GASTROENTEROLOGY Vol. 122, No. 4 the hepatic venous pressure gradient in patients with cirrhosis. J Hepatol 1990;11: Sabbá C, Ferraioli G, Genecin P, Colombato L, Buonamico P, Lerner E, Taylor KJW, Groszmann RJ. Evaluation of postprandial hyperemia in superior mesenteric artery and portal vein in healthy and cirrhotic humans: an operator-blind echo-doppler study. Hepatology 1991;13: O Brien S, Keogan M, Patchett S, McCormick PA, Afdhal N, Hegarty JE. Postprandial changes in portal hemodynamics in patients with cirrhosis. Gut 1992;33: Chou CC. Splanchnic and overall cardiovascular hemodynamics during eating and digestion. Fed Proc 1983;42: Brandt JL, Castleman L, Ruskin HD, Greenwald J, Kelly JJ, Jones A. The effect of oral protein and glucose feeding on splanchnic blood flow and oxygen utilization in normal and cirrhotic subjects. J Clin Invest 1955;34: Gallavan RH, Chou CC. Possible mechanisms for the initiation and maintenance of post-prandial intestinal hyperemia. Am J Physiol 1985;249:G301 G Krentz AJ, Boyle PJ, Macdonald LM, Schade DS. Octreotide: a long-acting inhibitor of endogenous hormone secretion for human metabolic investigations. Metabolism 1994;43: Polio J, Groszmann RJ. Hemodynamic factors involved in the development and rupture of esophageal varices: a pathophysiologic approach to treatment. Semin Liver Dis 1986;4: Hillon P, Blanchet L, Lebrec D. Effects of propranolol on hepatic blood flow in normal and portal hypertensive rats. Clin Sci 1982; 63: Kroeger RJ, Groszmann RJ. Effects of selective blockade of beta 2-adrenergic receptors on portal and systemic hemodynamics in a portal hypertensive rat model. Gastroenterology 1985;88: Alvarez D, De Las Heras M, Abecasis R, Terg R, Gerona S, Albornoz L, Galdame O, Torres J, Mastai R. Daily variation in portal blood flow and the effect of propranolol administration in a randomized study of patients with cirrhosis. Hepatology 1997; 25: Bandi JC, García-Pagán JC, Escorsell A, Francois E, Moitinho E, Rodés J, Bosch J. Effects of propranolol on the hepatic hemodynamic response to physical exercise in patients with cirrhosis. Hepatology 1998;28: Sabbá C, Ferraioli G, Buonamico P, Mahl T, Taylor KJW, Lerner E, Albano O, Groszmann RJ. A randomized study of propranolol on postprandial portal hyperemia in cirrhotic patients. Gastroenterology 1992;102: Brendtsen F, Simonsen L, Henriksen JH. Effect on hemodynamics of a liquid meal alone and in combination with propranolol in cirrhosis. Gastroenterology 1992;102: McCormack PA, Afdhal NH, Hegarty JE. Effects of propranolol on post-prandial increases in portal venous pressure in patients with portal hypertension (abstr). Gastroenterology 1987;92: Lamberts SWJ, Van Der Lely AJ, De Herder WW, Hofland LJ. Octreotide. N Engl J Med 1996;334: Hwang SJ, Lin HC, Chang CF, Lee FY, Lu CW, Hsia HC, Wang SS, Lee SD, Tsai YT, Lo KJ. A randomized controlled trial comparing octreotide and vasopressin in the control of acute esophageal variceal hemorrhage. J Hepatol 1992;16: McCormick PA, Biagini MR, Dick R, Greenslade L, Chin J, Cardin F, Wagstaff D, McIntyre N, Burroughs A. Octreotide inhibits the meal-induced increases in the portal vein pressure of cirrhotic patients with portal hypertension: a double-blind, placebo-controlled study. Hepatology 1992;16: Buonamico P, Sabba C, García-Tsao G, Berardi E, Antonica G, Ferraioli G, Jensen JE, Lerner E, Taylor KJW, Albano O, Groszmann RJ. Octreotide blunts portaprandial splanchnic hyperemia in cirrhotic patients: a double-blind randomized echo-doppler study. Hepatology 1995;21: Cooper AM, Braatvedt GD, Qamar MI, Brown H, Thomas DM, Halliwell M, Read AE, Corrall RJ. Fasting and post-prandial splanchnic blood flow is reduced by a somatostatin analogue (octreotide) in man. Clin Sci 1991;81: Albillos A, Rossi I, Iborra J, Lledó JL, Calleja JL, Barrios C, García P, Escartín P. Octreotide prevents postprandial splanchnic hyperemia in patients with portal hypertension. J Hepatol 1994;21: Malesci A, Tacconi M, Valentini A, Basilico M, Lorenzano E, Salerno F. Octreotide long-term treatment in patients with portal hypertension: persistent inhibition of postprandial glucagon response without major changes in renal function. J Hepatol 1997;26: Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60: Groszmann RJ, Glickman M, Blei AT, Conn HO, Storer E. Wedged and free hepatic venous pressure measured with a balloon catheter. Gastroenterology 1979;76: Chatila R, Ferayorni L, Gupta T, Groszmann RJ. Local arterial vasoconstriction induced by octreotide in patients with cirrhosis. Hepatology 2000;31: The North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. N Engl J Med 1988;319: Price BA, Jaffe BM, Zinner MJ. Effect of exogenous somatostatin infusion on gastrointestinal blood flow and hormones in the conscious dog. Gastroenterology 1985;88: Creutzfeldt W, Lembcke B, Fölsch UR, Schleser S. Effect of somatostatin analogue (SMS , Sandostatin) on pancreatic secretion in humans. Am J Med 1987;82: Silva G, Navasa M, Bosch J, Chesta J, Pizcueta P, Casamitjana R, Rivera F, Rodés J. Hemodynamic effects of glucagon in portal hypertension. Hepatology 1990;11: Okazaki K, Myazaky M, Onishy S, Ito K. Effects of food intake and various intrinsic hormones on portal blood flow in patients with liver cirrhosis demonstrated by pulsed Doppler with the Octoson. Scand J Gastroenterol 1986;21: Vatner SF, Franklin D, Van Citters RL. Mesenteric vasoactivity associated with eating and digestion in the conscious dog. Am J Physiol 1970;219: Wiest R, Tsai M, Groszmann RJ. Octreotide potentiates PKCdependent vasoconstriction in splanchnic arteries in portal-hypertensive and control rats. Gastroenterology 2001:120: Kroeger RJ, Groszmann RJ. Effect of selective blockade of 2 -adrenergic receptors on portal and systemic hemodynamics in a portal hypertensive rat model. Gastroenterology 1985;88: Escorsell A, Bandi JC, Andreu V, Moitinho E, García-Pagan JC, Bosch J, Rodés J. Desensitization to the effects of intravenous octreotide in cirrhotic patients with portal hypertension. Gastroenterology 2001;120: Fagan TC, Sawyer PR, Gourley LA, Lee JT, Gaffney TE. Postprandial alterations in hemodynamics and blood pressure in normal subjects. Am J Cardiol 1986;58: Kelbaek H, Munck O, Christensen NJ, Godtfredsen J. Central hemodynamic changes after a meal. Br Heart J 1989;61: Received March 19, Accepted November 28, Address requests for reprints to: Julio Vorobioff, M.D., Liver Unit and Hepatic Hemodynamic Laboratory, Sanatorio Parque & Fundación Dr. J. R. Villavicencio, Bvrd. Oroño 860, 2000 Rosario, Argentina. vorodiez@intramed.net.ar; fax: (54) The authors acknowledge Professor Roberto Groszmann, M.D., for critical advice and insightful discussions. The authors offer special thanks to Novartis Pharmaceuticals (U.S.) for octreotide serum level measurements and to Alicia Santaliestra, M.D. (Novartis Argentina), for her generous assistance.