Portal Hyperperfusion after Extended Hepatectomy Does Not Induce a Hepatic Arterial Buffer Response (HABR) but Impairs Mitochondrial Redox State and Hepatocellular Oxygenation

Background & Aims

Portal hyperperfusion after extended hepatectomy or small-for-size liver transplantation may induce organ dysfunction and failure. The underlying mechanisms, however, are still not completely understood. Herein, we analysed whether hepatectomy-associated portal hyperperfusion induces a hepatic arterial buffer response, i.e., an adaptive hepatic arterial constriction, which may cause hepatocellular hypoxia and organ dysfunction.

Methods

Sprague-Dawley rats underwent 30%, 70% and 90% hepatectomy. Baseline measurements before hepatectomy served as controls. Hepatic arterial and portal venous flows were analysed by ultrasonic flow measurement. Microvascular blood flow and mitochondrial redox state were determined by intravital fluorescence microscopy. Hepatic tissue pO2 was analysed by polarographic techniques. Hepatic function and integrity were studied by bromosulfophthalein bile excretion and liver histology.

Results

Portal blood flow was 2- to 4-fold increased after 70% and 90% hepatectomy. This, however, did not provoke a hepatic arterial buffer response. Nonetheless, portal hyperperfusion and constant hepatic arterial blood flow were associated with a reduced mitochondrial redox state and a decreased hepatic tissue pO2 after 70% and 90% hepatectomy. Microvascular blood flow increased significantly after hepatectomy and functional sinusoidal density was found only slightly reduced. Major hepatectomy further induced a 2- to 3-fold increase of bile flow. This was associated with a 2-fold increase of bromosulfophthalein excretion.

Conclusions

Portal hyperperfusion after extended hepatectomy does not induce a hepatic arterial buffer response but reduces mitochondrial redox state and hepatocellular oxygenation. This is not due to a deterioration of microvascular perfusion, but rather due to a relative hypermetabolism of the remnant liver after major resection.     

Method for measuring hepatic uptake of oxygen or other blood-borne substances in situ.

A preparation is described by which hepatic arterial blood flow and portal venous blood flow can be accurately and continuously measured while simultaneously providing a method by which multiple blood samples can be taken from the hepatic artery, portal vein, and hepatic vein without disrupting hepatic hemodynamics or causing hemodilution. By this means hepatic uptake or release of blood-borne substances can be measured in situ and correlated with hemodynamic parameters. In 13 splenectomized cats, oxygen uptake by the denervated liver was 4.5 +/- 0.3 ml . min-1. 100 g-1 of tissue, representing 54% of total oxygen removed by the splanchnic bed. The hepatic hemodynamics determined by this method are similar to those reported by others in vivo and the metabolic state of the liver remained stable for at least 2 h during which an average of 29 blood samples were taken. Advantages of this preparation over other methods of obtaining similar data are discussed.     

Effect of hypoxemia on the renin-angiotensin-aldosterone system in humans

Hypoxemia was induced in five subjects older than 40 (group 1) and five younger than 35 yr (group 2) on normal and low-salt diets by having the subjects breathe hypoxic gas. The fractional inspired O2 of the hypoxic gas was regulated so that group 1 hemoglobin saturations fell to 90% for 1 h. Group 2 subjects had desaturation to 90% for 1 h followed by desaturation to 80% for a 2nd h. Plasma renin activity (PRA), angiotensin-converting enzyme activity (ACE), and plasma cortisol levels did not change during hypoxemia. Plasma aldosterone levels fell in both groups during the 1st h of hypoxemia. Decreases were greatest during salt restriction and were significant (P less than 0.01) for the combined groups. Plasma aldosterone levels plateaued during the 2nd h of more severe hypoxemia in group 2. Hepatic blood flow, measured by indocyanine green clearance, and the adrenal response to exogenous adrenocorticotropic hormone, measured by changes in plasma cortisol and aldosterone, were not changed by hypoxemia in group 2 subjects. These results indicate that plasma aldosterone falls during hypoxemia despite unchanged PRA, ACE, hepatic blood flow, and adrenal function.     

Transport of amino acids in the splanchnic bed by blood plasma and blood in the preruminant calf]

Three preruminant calves were fitted with catheters in portal and hepatic veins and in a mesenteric artery. Two electromagnetic flowmeter probes were clipped around the portal vein and the hepatic artery. The calves were fed either a diet with a low (L) or a high (R) abomasal emptying rate for dietary proteins. Blood flow and free amino acid levels in plasma (P) and blood (S) were determined before the morning meal and during the following 7 h. In the portal vein, for most amino acids P/S ratios were correlated to the net amino acid balance of the digestive tract measured in plasma. By contrast in the hepatic vein, these ratios were mainly correlated to hepatic balance measured in whole blood. Correlations between digestive tract and hepatic balance calculated using either plasma or whole blood pool were different for some amino acids. This suggests that amino acid exchange between plasma and blood cells is low and absorbed amino acids are mainly transported to the liver by plasma, whereas whole blood rather than plasma is concerned in amino acid exchanges in the liver.     

Role of afferent nerves and sensory peptides in the mediation of hepatic artery buffer response.

Intrahepatic arteries are richly innervated by both adrenergic and sensory vanilloid-sensitive (capsaicin-sensitive) fibers. Stimulation of capsaicin sensitive fibers has been shown to dilate the intrahepatic vessels by both releasing sensory neuropeptides and by modulating the adrenergic tone. However the participation of capsaicin-sensitive fibers in the mediation of the hepatic artery buffer response (HABR) has not been investigated yet. To explore the involvement of sensory innervation and sensory neuropeptides in the HABR, the experiments were performed on capsaicin-denervated Wistar rats. In addition, we used selective CGRP and tachykinin receptor antagonists to test the participation of CGRP, substance P and NK-A in HABR in the rat. In anesthetized rats the hepatic artery blood flow (HABF), microcirculatory hepatic blood flow (HBF) and portal blood flow (PBF) were determined. The HABR was induced by partial occlusion of the portal vein and maintaining the PBF at 10% of its control preocclusive value. In the control HABR the hepatic artery blood flow increased by 89% (p< 0.005) whilst the HBF at the same time decreased by 32% (p< 0.005) in comparison to preocclusive HABF and HBF values. In sensory-denervated rats the resting HBF and PBF were increased by 23% (p< 0.05) and 34% (p< 0.05), respectively in comparison to the control HBF and PBF values. In this group the induction of the HABR increased the hepatic artery blood flow by only 55% (p< 0.05), whilst the HBF was reduced by 45% (p< 0.05). Pretreatment with CGRP 8-37 (CGRP receptor antagonist) and NK-1 but not NK-2 nor NK-3 receptor antagonists significantly reduced the HABF by 43% (p< 0.05) and 25% (p< 0.05) as compared to the HABF value in the control HABR group. These findings support the hypothesis that the hepatic artery buffer response induced by reduction of the portal inflow to the liver by 90% is partially mediated by activation of capsaicin-sensitive sensory fibers in the liver, probably due to local tissue ischemia and hypoxia. The observed vasodilation in the vascular bed of the hepatic artery is due to stimulation of CGRP and NK-1 receptors.     

Dissociation of adrenal androgen and cortisol levels in acute stress

Patients recovering from acute surgical stress often excrete increased 17-OH corticosteroids with no change in 17-ketosteroids. The explanation for these findings is unclear. In order to investigate possible divergence between cortisol and adrenal androgen metabolism in acute stress, repeated morning cortisol and dehydroepiandrosterone (DHA) measurements were made in patients undergoing ACTH stimulation 48 to 96 hours preoperatively, followed by determinations before and during major surgery, also performed in the morning. Cortisol and DHA are largely metabolized by the liver, so liver blood flow under a constant general anesthetic regimen known not to affect cortisol metabolism was monitored by pre- and intraoperative indocyanine green dye clearance. Results indicated no difference between the cortisol and DHA stimulation resulting from two hours of ACTH stimulation or major surgery, and a small (14.4%) decline in hepatic blood flow during general anesthesia. However, while DHA concentrations remained constant immediately preceding surgery, cortisol concentrations increased by 61% (P less than 0.05). Previous studies have also demonstrated increased concentrations of cortisol before surgical procedures, presumably due to psychological stress. However, this is the first demonstration of a dissociation between concentrations of cortisol and an adrenal androgen due to psychological stress.     

Nycthemeral Changes in Plasma Cortisol Levels in Preruminant Calves with Different Milk Proteins

Preruminant calves bearing indwelling catheters in the hepatic artery, the portal and the hepatic veins were fed with two kinds of diets, a conventional curdled milk diet and a milk diet which was uncurdled in the ahomasum. Measurements of plasma cortisol in blood sampled regularly during the 24 hr of the day indicated that with curdled milk, cortisol concentrations were significantly higher than with uncurdled milk. Nycthemeral changes were characterized by high values before meals and by postprandial decreases. Between meals, several peak values were observed and in the night a regular increase occurred. With both kinds of meals, cortisol evolutions were similar though peak values were higher with the curdled milk.     

Nycthemeral Changes in Plasma Cortisol Levels in Preruminant Calves with Different Milk Proteins

Preruminant calves bearing indwelling catheters in the hepatic artery, the portal and the hepatic veins were fed with two kinds of diets, a conventional curdled milk diet and a milk diet which was uncurdled in the ahomasum. Measurements of plasma cortisol in blood sampled regularly during the 24 hr of the day indicated that with curdled milk, cortisol concentrations were significantly higher than with uncurdled milk. Nycthemeral changes were characterized by high values before meals and by postprandial decreases. Between meals, several peak values were observed and in the night a regular increase occurred. With both kinds of meals, cortisol evolutions were similar though peak values were higher with the curdled milk.     

不同入肝血流阻断方式对荷瘤小鼠肝功能影响的实验研究

目的:探讨不同血流阻断方式对荷瘤小鼠肝细胞功能的影响。方法:选择昆明小鼠24只随机分为三组,正常对照组(Suspe-nded operation,SO)、肝门阻断组(Occlusion of the portal triad,OPT)、保留肝动脉持续阻断门静脉(Occlusion of portal vein,OPV)各8只。采用门静脉注射肿瘤的方法建立肝癌模型,建模后3天采用阻断范围为左外叶和中叶、阻断时间为60分钟的入肝血流阻断方式,复流后5天后,通过测量3组对肝脏的缺血再灌注损伤程度以及病理学变化来评价不同血流阻断方式对肝细胞功能影响的程度。结果:门静脉注射小鼠肝癌细胞8天后,对照组测量小鼠正常丙氨酸氨基转移酶(ALT)值为66.5±22.3 IU/L,OPT组值为276.3±80.5 IU/L,OPV组值为89.6±28.4 IU/L,两组比较有统计学差异(P0.01);对照组测量小鼠正常天冬氨酸氨基转移酶(AST)值为301.3±126.7 IU/L,OPT组值为1126.4±285.5 IU/L,OPV组值为438.6±150.7 IU/L,两组比较有统计学差异(P0.01),病理组织学OPV组肝细胞损伤程度明显较OPT组轻。结论:保留肝动脉持续阻断门静脉可以减轻荷瘤小鼠肝脏的缺血再灌注损伤。     

Simultaneous sampling of portal, hepatic and systemic blood during intragastric loading and tracer infusion in conscious pigs

A surgical model for catheterization at multiple sites has been developed for use in long-term metabolic studies. For blood sampling, catheters were inserted into the portal and hepatic veins and the common carotid artery. The hepatic vein catheter was inserted from the margin of a liver lobe and led through the venous system, until the tip was close to the bifurcation with the inferior vena cava. A new technique was developed to ensure correct placement of the hepatic vein catheter using the specific extraction of indocyanin-green over the liver during surgery. Gastrostomy was performed using a Pezzer catheter. Catheters in the artery and hepatic and portal veins were patent for blood withdrawal for up to 4 weeks, and thus allowed repeated metabolic studies. Studies were performed in conscious animals familiar with the experimental situation.     

ACTH inhibits the development of primary decompensation of the systemic and portal circulation in acute blood loss in rats]

Using the method of contact luminescent biomicroscopy of the liver and intestine coupled with the measurement of systemic blood pressure by micromanometer and ultrasonic registration of blood flow velocity in portal vein and hepatic artery it has been established that in rats with acute decompensatory hemorrhage fragments of ACTH (1-24) and (4-10) improve the state of portal macro- and microcirculation and increase the life span 2-3-fold. ACTH does not influence the dynamics of acute compensatory hemorrhage and the development of the posthemorrhagic microcirculatory disturbances (local microstases, microthromboses, erythrocyte aggregation).     

Combined Hepatic Vein, Umbilicoportal Vein, and Superior Mesenteric Artery Catheterization in Portal Hypertension: Estimation of the Portal Fraction of Total Hepatic Blood Flow in Cirrhotic Patients

Hemodynamic data were obtained in 13 cirrhotic patients with severe portal hypertension, undergoing combined hepatic vein, umbilicoportal vein, and superior mesenteric artery catheterization. The relative clearance of indocyanine green, the portohepatic gradient (difference between the free portal venous pressure and the free hepatic venous pressure), and the estimated hepatic blood flow were measured. The portal fraction (PF) of total hepatic blood flow was calculated in all patients using indicator dilution curves obtained from the portal bifurcation, a right hepatic vein, and when possible a left hepatic vein (six cases) after injection of ⁵¹Cr-labeled red blood cells (⁵¹Cr RBC) into the superior mesenteric artery. Flows were overestimated because of loss of indicator through spontaneous portosystemic shunts; however, the ratio between hepatic and portal indicator dilution curves can be used to calculate the portal fraction of total hepatic blood flow since no extrahepatic shunts existed after the bifurcation of the portal vein (as shown on portography). In 10 patients, 15 series of curves were calculable and the PF varied between 30.1 and 100% (mean ± SE: 71.1 ± 6.2%). In the three other patients, only delayed activity from recirculation was detected from portal and hepatic vein samples and PF was 0%; in these three cases, portography and arteriography revealed spontaneous portacaval shunting with reverse and/or stagnant circulation in the portal vein. In the 13 patients, no correlation existed between PF and the relative clearance of indocyanine green or the portohepatic gradient, parameters generally used as indices of severity in cirrhosis. In 10 patients, no correlation was found between PF and the estimated hepatic blood flow.     

Hepatic neovascularization after partial portal vein ligation: novel mechanism of chronic regulation of blood flow

The present study was undertaken to investigate hepatic microcirculatory response following partial portal vein ligation (PPVL) in rats. Portal pressure was markedly increased 2-6 wk after PPVL, but no significant reduction in sinusoidal perfusion and hepatocellular injury were detected. However, marked neovascularization was observed in PPVL rats using intravital microscopy and scanning electron microscopy (SEM). Extremely high red blood cell velocity (2,000-4,900 microm/s) was seen in these vessels. Injection of fluorescein sodium via the carotid artery revealed that the neovessels originated from the hepatic arterial vasculature. This was further confirmed by clamping the common hepatic artery and phenylephrine injection from the carotid artery. These vessels maintained sufficient flow after massive sinusoidal shutdown elicited by the portal infusion of endothelin receptor B agonist IRL-1620. SEM also showed extensive neovascularization at the hilum. Additionally, clamping the portal vein decreased sinusoidal perfusion only by 9.5% in PPVL, whereas a 71.2% decrease was observed in sham. These results strongly suggest that the liver maintains its microcirculatory flow by vascular remodeling from the hepatic arterial vasculature following PPVL.     

Different accessibility from the artery and the portal vein of alpha- and beta-receptors involved in the sympathetic nerve action on glycogenolysis and hemodynamics in perfused rat liver

In perfused rat liver perivascular nerve stimulation (7.5 Hz, 20 V, 2 ms, 5 min) at the liver hilus caused an increase in glucose and lactate output and a decrease in flow. The influence of the alpha 1-receptor blocker prazosine and the beta-blocker propranolol on these nerve effects was studied in the isolated rat liver perfused classically via the portal vein only and, as developed recently, via both the hepatic artery and the portal vein. 1) In livers perfused via the portal vein only the nerve stimulation-dependent metabolic alterations were nearly completely inhibited by prazosine (5 microM), but not influenced by propranolol (10 microM). The hemodynamic changes were lowered to only 33% by prazosine and not altered by propranolol either. 2) In livers perfused via the hepatic artery (100 mm Hg, 20-40% of flow) and the portal vein (10 mm Hg, 80-60% of flow)--similar to portal perfusions--the nerve stimulation--dependent metabolic alterations were almost completely blocked by arterial, portal or simultaneously applied arterial and portal prazosine. However--in contrast to portal perfusions--the metabolic alterations were reduced to about 20% (glucose) and 50% (lactate) also by propranolol independently of its site of application. The decrease in flow was reduced by prazosine to about 60%, 50% and 30% when applied via the artery, the portal vein or via both vessels, respectively. The hemodynamic alterations were not influenced by propranolol. These results allow the following conclusions: A subpopulation of beta-receptors can play a permissive role in the alpha 1-receptor-mediated sympathetic nerve action on glucose and lactate metabolism.(ABSTRACT TRUNCATED AT 250 WORDS)     

The use of 8-phenyltheophylline as a competitive antagonist of adenosine and an inhibitor of the intrinsic regulatory mechanism of the hepatic artery

Reduction of portal blood flow results in compensatory vasodilation of the hepatic artery, the hepatic arterial buffer response. The hypothesis tested is that the regulation of the buffer response is mediated by adenosine, where the local concentration of adenosine in the region of the hepatic arterial resistance vessels is regulated by washout of adenosine into portal venules that are in intimate contact with hepatic arterioles. In anesthetized cats, portal flow was reduced to zero by complete occlusion of all arterial supply to the guts. The resultant dilation of the hepatic artery compensated for 23.9 +/- 4.9% of the decrease in portal flow. Dose-response curves were obtained for the effect of intraportal adenosine infusion on hepatic arterial conductance in doses that did not lead to recirculation and secondary effects on the hepatic artery via altered portal blood flow. The dose to produce one-half maximal response for adenosine is 0.19 mg X kg-1 X min-1 (intraportal) and the estimated maximal dilation is equivalent to an increase in hepatic arterial conductance to 245% of the basal (100%) level. The adenosine antagonist, 8-phenyltheophylline, produced dose-related competitive antagonism of the dilator response to infused adenosine (but not to isoproterenol) and a similar, parallel antagonism of the hepatic arterial buffer response. If supramaximal blocking doses were used, the hepatic artery showed massive and prolonged constriction with blood flow decreasing to zero. The data strongly support the hypothesis that intrinsic hepatic arterial buffer response is mediated entirely by local adenosine concentration.(ABSTRACT TRUNCATED AT 250 WORDS)     

肝硬化患者肝脏门静脉血流与肝叶萎缩之间关系的研究

目的：探讨肝硬化患者肝脏右叶、左叶体积变化,检测肝硬化患者门静脉血流情况,分析二者之间的关系,以及门静脉血流与肝功能之间关系。方法：本研究纳入54例肝硬化患者和40例正常人,采用超声多普勒方法分析这些受试者的肝脏体积和门静脉主干及左右分支的内径、血流速、流量数据,并通过静脉血检测白蛋白、胆红素、胆碱酯酶水平等评估患者肝功能水平。结果：肝硬化组平均年龄46.3岁,男性32例,其中childA级患者16例,childB级患者27例,childC级患者11例;正常对照组平均年龄41．8岁,男性24例。肝硬化组患者右左肝叶之比明显低于正常对照组（p〈0．05）,门静脉内径和血流量明显高于正常对照组（p〈0．05）．随着child分级升高,门静脉血流量也明显升高。肝硬化组门静脉右支血流量明显低于左支血流量（p〈0．05）;此外肝硬化患者门静脉右支和左支血流量之比明显低于正常人群门静脉右左支之比（p〈0．05）;而且肝硬化患者门静脉右左支血流量之比与右左肝叶具有明显的相关性与右左肝叶之比具有明显的相关性（r=0．64,p〈0．05）。结论：评估肝硬化病人门静脉血流情况,对于判断肝脏病理变化程度,评价治疗效果,以及选择治疗方案方面都具有重要的临床价值     

The blood supply to the abdominal organs of the fetal guinea-pig

Guinea-pigs near term of pregnancy were anaesthetized with diazepam and sodium pentobarbitone. A fetus was exposed and the vitelline artery catheterized to measure blood pressure and heart rate or to render a reference sample of blood for the determination of organ blood flow by the microsphere technique. The radioactive microspheres were injected through a catheter in the right atrium. Mean arterial blood pressure was 4.0 kPa and heart rate was 261 beats min-1. The liver, spleen, pancreas and gut receive most of their blood supply from the same trunk as the vitelline artery. The sample from this vessel was also used to calculate blood flow to the adrenal glands, kidneys, urogenital tract, and placenta, assuming even mixing of microspheres and blood in the abdominal aorta. Umbilical blood flow, corrected to a fetal weight of 100 g, averaged 7.5 ml min-1. The adrenal glands, which are known to increase their cortisol secretion near term, had a very high rate of perfusion. If the microspheres were injected in the umbilical vein, almost all were trapped in the liver, confirming the absence of a ductus venosus in the guinea-pig fetus. Most of these microspheres were found in the quadrate lobe of the liver. Hepatic arterial blood flow was also unequally distributed, with more than two-thirds going to the right lobe of the liver. Although the distribution of portal venous blood flow is not known, it is evident that different areas of the liver are presented with blood of greatly varying oxygen saturation.     

The effect of pancreatic and intestinal venous blood on hepatic atrophy and compensatory hyperplasia in the rat

Hepatotrophic effect of pancreatic and intestinal venous blood was studied in rats with mesocaval or distal splenocaval shunt following ligation of a branch of the portal vein supplying 70% of liver mass. Because 2/3 of liver mass was deprived of portal flow the nonligated liver lobes were not hypoperfused due to shunt procedure. During the first three postoperative days the DNA synthesis, mitotic index, and changes in relative weights were measured in both ligated (atrophied) and nonligated (compensatory hyperplasia) parts of the liver. It was found, that the restorative capacity of the liver existed in rats with selective portasystemic shunts. The stimulus to growth was greater in lobes supplied by intestinal venous blood compared to those perfused by pancreatic effluent. The increase in DNA synthesis occurred in lobes undergoing atrophy and the intensity of this response was also dependent on type of shunt since recirculation of intestinal blood by way of the hepatic artery inhibited atrophy to a greater extent than pancreatic venous effluent. Although the patency of arterial branches was confirmed the ligated lobes showed necrotic lesions. Systemic recirculation of intestinal venous blood far more inhibited necrosis than pancreatic venous blood.     

A portal-arterial glucose concentration gradient as a signal for an insulin-dependent net glucose uptake in perfused rat liver

Since in the usual perfusion of isolated rat liver via the portal vein an insulin-dependent increase of hepatic glucose uptake could not be demonstrated, the possibility was considered that hepatic glucose uptake might not be a function of the absolute concentration of this substrate but of its concentration gradient between the portal vein and the hepatic artery. Therefore a new method was established for the simultaneous perfusion of isolated rat liver via both the hepatic artery (20-35% flow) and the portal vein (80-65% flow). When glucose was offered in a concentration gradient, 9.5 mM in the portal vein and 6 mM in the hepatic artery, insulin given via both vessels caused a shift from net glucose release to uptake. This insulin-dependent shift was not observed when glucose was offered without a gradient or with an inverse gradient, 6 mM in the portal vein and 9.5 mM in the hepatic artery. Using a portal-arterial glucose gradient as a signal the liver might be able to differentiate between endogenous and exogenous glucose.     

Whole blood and plasma amino acid transfers across the portal drained viscera and liver of the pig.

Whole blood (WB) and plasma (P) amino acid transfers across the portal drained viscera and the liver were determined during 6 h of a constant p-aminohippuric acid infusion in three hourly-fed Landrace x Large-White pigs (30.5 kg, mean live weight) surgically prepared with chronically inserted catheters in a mesenteric vein (MV), the portal vein (PV), an hepatic vein (HV) and the carotid artery (CA). Plasma and WB amino acid concentrations were determined in the CA, PV and HV. The plasma/WB ratios showed no significant differences for vessels except for lysine and glutamate for which this ratio is significantly higher in the HV and in the PV for lysine. This suggests that the PV lysine and HV glutamate were preferentially transported in the plasma. In the PV, threonine, valine and alanine are transported by both plasma and red blood cells. These data show that the contribution of plasma and whole blood to amino acid transport can be different between amino acids and between individual tissues.