Samuel S. Lee

Abnormal tissue oxygenation in patients with cirrhosis and liver failure

Systemic haemodynamic and hepatic venous pressures, arterial and mixed venous gases and arterial lactate concentration were measured in 35 patients with histologically proven alcoholic cirrhosis who had been classified into three groups (A, B and C). Eight alcoholic patients without cirrhosis on liver biopsy were also studied. Compared with group A patients, group C patients had significantly higher hepatic venous pressure gradient, cardiac index, O2 transport and arterial lactate concentration and significantly lower systemic vascular resistance, arteriovenous O2 content difference and O2 uptake. In group B patients, corresponding values fell between those of groups A and C. Group A patients, unlike group C patients, were not significantly different from patients without cirrhosis with respect to cardiac index, systemic vascular resistance, O2 uptake and arterial lactate concentration. Our results suggest that in patients with cirrhosis, liver failure-associated hyperdynamic circulation may be accompanied by an abnormal tissue oxygenation.

Reduction of Intrahepatic Vascular Space in the Pathogenesis of Portal Hypertension In Vitro and In Vivo Studies in the Rat

To elucidate a possible role for reduction of intrahepatic vascular space in the pathogenesis of portal hypertension, we studied a partially hepatectomized rat model in vitro and in vivo. The in vitro study used livers from normal, 1/3-hepatectomized, and 2/3-hepatectomized rats, and rats with cirrhosis caused by chronic bile duct ligation, for isolated, perfused flow-pressure plotting. Resistance to perfusion increased such that significant differences were found between all groups except the last two, which showed similar resistances. The in vivo study measured splanchnic blood flow by radioactive microspheres and portal pressure in anesthetized sham-operated and 1/3- and 2/3-hepatectomized rats. Although absolute portal tributary blood flows did not change, portal flow per gram of remnant liver showed significant increases: 1.57 +/- 0.32 ml/min X g liver, 2.52 +/- 0.60 ml/min X g liver, p less than 0.01; 3.48 +/- 1.04 ml/min X g liver, p less than 0.01, respectively. Although intrahepatic resistance increased significantly only in the 2/3-hepatectomized group, portal pressures increased significantly in both groups of hepatectomized rats: normal, 7.5 +/- 1.1 mmHg; 1/3-hepatectomized, 9.4 +/- 1.1 mmHg; and 2/3-hepatectomized, 11.1 +/- 1.2 mmHg. Thus, decreased intrahepatic vascular space caused by resection and hepatocellular hypertrophy leads to portal hypertension, thereby suggesting that this reduction in space may be the pathogenic factor common to a number of different theories of portal hypertension.

Circulatory effects of somatostatin analogue in two conscious rat models of portal hypertension.

A somatostatin analogue, a long-acting octapeptide (SMS 201-995), has been reported to decrease portal pressure, but the mechanism is unclear. To elucidate the effects of this drug on both systemic and splanchnic hemodynamics, it was administered in two conscious rat models of portal hypertension. The dose-response curves showed that the somatostatin analogue significantly decreased portal pressure at a lower dose in rats with cirrhosis than in portal vein-stenosed rats. Calculated ED50 values were significantly different among all groups. Intravenous infusion of 8 micrograms/kg body wt.h of somatostatin analogue significantly decreased cardiac output by approximately 20% in both groups of portal hypertensive rats and increased mean arterial pressure by 7%. Accordingly, systemic vascular resistance markedly increased, indicating vasoconstrictor effects of this drug. The somatostatin analogue also significantly decreased portal tributary blood flow by 18% in portal vein-stenosed rats and 27% in cirrhotic rats. In sham-operated rats, somatostatin analogue had no effect on the systemic or splanchnic circulation. This study shows that somatostatin analogue decreases portal pressure principally by reducing portal tributary blood flow. This reduction may be due to either a direct vasoconstrictive effect or diminution in vasoactive hormone release.

Possible deleterious hemodynamic effect of nifedipine on portal hypertension in patients with cirrhosis

The acute effects of nifedipine, 10 mg administered sublingually, were studied in 10 patients with alcoholic cirrhosis. Nifedipine significantly increased cardiac output and reduced systemic vascular resistance. Nifedipine also increased the hepatic venous pressure gradient by 15% (P < 0.01). Hepatic blood flow and azygos blood flow did not change significantly. It is suggested that nifedipine increases portal pressure and thus may be deleterious to patients with portal hypertension.

Regional Blood Flows by the Microsphere Method: Reproducibility in Portal Hypertensive Rats and Influence of a Portal Vein Catheter

Abstract To determine the reproducibility of splanchnic blood flow measurements by the microsphere method in rats with portal hypertension and the effects of laparotomy with portal vein cannulation, eight groups of 10 rats were studied. Cardiac output and regional blood flows were measured twice, 10 min apart, in pentobarbital anesthetized or awake, sham-operated or portal vein-ligated rats, with or without portal cannulation. Variability between the two successive measurements was not affected by portal hypertension or portal cannulation, and was not different in the splanchnic territory and in other organs. Laparotomy with portal cannulation had no significant effect in sham-operated rats. In awake portal hypertensive rats, cardiac output (53.9 ± 3.0 vs 45.8 ±2.9 ml · min-1 · 100 g body wt-l, P < 0.01) and splanchnic blood flow (12.31 ± 0.72 vs 9.34 ± 0.85 ml · min-1 · 100 g body wt-1, P < 0.01) were lower in portal vein cannulated rats compared with those of non-cannulated animals. In anesthetized portal hypertensive rats blood flows were unaffected by portal cannulation, but arterial pressure (100.2 ± 4.3 vs 119.9 ± 3.4 mm Hg, P < 0.01) and heart rate (366.5 ± 10.0 vs 405.5 ± 7.4 beats · min-1, P < 0.01) were elevated. Anesthesia also decreased portal pressure (14.8 ± 0.5 vs 12.0 ± 0.4 mm Hg, P < 0.05) in portal hypertensive rats. We conclude that the microsphere method remains reproducible in portal hypertensive rat models. Laparotomy with portal cannulation can alter systemic and splanchnic hemodynamics in portal hypertensive rats; these effects can also be changed during pentobarbital anesthesia. Regional blood flow measurements in portal hypertensive rats should be performed in animals without portal cannulation and preferably in the awake state.

Reversal of adrenaline-induced increase in azygos blood flow in patients with cirrhosis receiving propranolol

In patients with cirrhosis, endogenous catecholamines may influence the circulatory effects of propranolol. We intended to evaluate the interaction of adrenaline and propranolol on azygos blood flow, an estimate of blood flow in the superior portosystemic collateral circulation. We investigated 6 patients with cirrhosis, 5 with good liver function, receiving an intravenous infusion of adrenaline (50 ng/kg/min) before and after administration of propranolol. The median value for baseline azygos blood flow was increased from 700 ml/min (range 340-1 470 ml/min) to 1 050 (range 570-1 840 ml/min) with adrenaline alone (P less than 0.05), and decreased to 610 ml/min (range 260-1 190 ml/min) with propranolol alone (P less than 0.05). The infusion of adrenaline given after propranolol further reduced azygos blood flow to a median value of 530 ml/min (range 200-730 ml/min) (P less than 0.05). Thus, following beta-adrenergic blockade, there is a reversal of the effects of adrenaline on azygos blood flow, which corresponds to a potentiation of the effects of propranolol. Similar endogenous adrenaline-propranolol interactions may play a role in preventing recurrent variceal bleeding in cirrhotic patients.

Comparative haemodynamic effects of betaxolol and propranolol in patients with cirrhosis

The acute effects of betaxolol (10 mg, intravenously), a new cardioselective beta-blocker, and propranolol (15 mg, intravenously) on splanchnic and systemic circulations were studied in two matched groups of six patients with portal hypertension due to cirrhosis. Similar decreases in hepatic venous pressure gradient and azygous blood flow--an estimation of superior portosystemic shunts--were observed after both drugs, whereas hepatic blood flow was not modified. The decreases in heart rate and cardiac index were also similar after betaxolol and propranolol. Both drugs induced a significant decrease in the fraction of cardiac output flowing through superior portosystemic shunts. These findings confirm that the marked effect of beta-adrenoceptor blocking agents on splanchnic circulation results both from the reduction in cardiac output and from a vasoconstriction of the portal vein territory, and demonstrate that this vasoconstriction of the portal vein area does not necessitate a beta 2-blocking activity of the drug. The similar efficiency of the two agents in decreasing the hyperkinetic circulation suggests that betaxolol merits further long-term study in the pharmacologic treatment of portal hypertension.

Oxygen and bile acid content in the azygos blood: Clues to the azygos derivation in patients with portal hypertension

In patients with cirrhosis, elevation of azygos blood flow has been attributed on indirect grounds to cephalad portosystemic collaterals. To gather more information on the origin of the azygos blood, we studied the oxygen and bile acid content of the azygos and mixed venous blood in patients with portal hypertension. Azygos oxygen saturation was 59.6 +/- 6.0% in 8 controls, and significantly higher in 35 patients with cirrhosis (76.7 +/- 7.6%; P less than 0.01) as well as in 6 patients with noncirrhotic portal hypertension (84.0 +/- 8.2%; P less than 0.01). High oxygen saturation, however, was not correlated to azygos blood flow in patients with cirrhosis. In cirrhotic patients, total bile acid concentrations were 28.1 +/- 20.4 mumol/l in the pulmonary artery and 25.9 +/- 17.6 mumol/l in the azygos vein, giving an azygos to mixed venous ratio of 0.95 +/- 0.18. These results provide new evidence that elevated azygos blood flow in patients with portal hypertension is derived from the portal system, and perhaps predominantly from the splenic territory.

Haemodynamic rebound phenomena after abrupt cessation of propranolol therapy in portal hypertensive rats

The haemodynamic effect of sudden termination of propranolol therapy was studied in sham-operated and portal hypertensive rats. All animals were injected with propranolol (20 mg/kg/day) or saline i.p. for 10 days, then had an isoproterenol infusion test performed 48 h or 72 h after cessation of injections. The dose of isoproterenol required to increase the heart rate by 50 beats/min (CD50), was significantly lower in both sham-operated and portal hypertensive rats at 48 h after propranolol withdrawal. Maximum chronotropic response (Rmax), was significantly higher only in portal hypertensive rats at 48 h after propranolol withdrawal. These results show the existence of a transient beta-adrenergic hypersensitivity state following propranolol withdrawal in normal and portal hypertensive rats.

Divergent circulatory effects of betaxolol in conscious and anesthetized normal and portal hypertensive rats

We aimed to define the circulatory effects of beta 1-blockade in conscious normal and portal hypertensive rats and determine if pentobarbital anesthesia affected these responses. A selective beta 1-antagonist, betaxolol, was given to four groups: conscious and anesthetized sham-operated and portal hypertensive rats. Cardiac output and splanchnic organ blood flows were measured by radioactive microspheres twice in each rat, before and 15 min after betaxolol. Both groups of conscious rats maintained mean arterial pressure despite significant decreases in cardiac output and heart rate, by increasing total peripheral resistance. Anesthetized rats were unable to do this and thus also diminished arterial pressure significantly, with portal hypertensive rats showing greater decreases than sham-operated rats. Portal tributary flow and portal pressure decreased only in the anesthetized rats. Autoregulation of splanchnic blood flow was not uniform between groups or organs: although splenic flow decreased in all four groups, intestinal blood flow decreased only in anesthetized portal hypertensive rats. The greatest decreases in several splanchnic organ blood flows were seen in this latter group. These results indicate that: (i) pentobarbital markedly changes systemic and splanchnic responses to beta 1-blockade; (ii) splanchnic autoregulation is not uniform--the intestinal circulation enjoys more protection than the splenic; and (iii) portal hypertensive rats seem to be more vulnerable to the circulatory effects of beta 1-blockade.