Hyperglucagonemia in Laennec's cirrhosis. The role of portal-systemic shunting

Unraveling the mechanisms of hepatogenous diabetes and its therapeutic perspectives

The review focused mainly on the pathogenesis of hepatogenous diabetes (HD) in liver cirrhosis (LC). This review reveals parallels between the mechanisms of metabolic dysfunction observed in LC and type II diabetes (T2DM), suggesting a shared pathway leading to HD. It underscores the role of insulin in HD pathogenesis, highlighting key factors such as insulin signaling, glucose metabolism, insulin resistance (IR), and the influence of adipocytes. Furthermore, the impact of adipose tissue accumulation, fatty acid metabolism, and pro-inflammatory cytokines like Tumor necrosis factor-α (TNF-α) on IR are discussed in the context of HD. Altered signaling pathways, disruptions in the endocrine system, liver inflammation, changes in muscle mass and composition, and modifications to the gut microbiota collectively contribute to the complex interplay linking cirrhosis and HD. This study highlights how important it is to identify and treat this complex condition in cirrhotic patients by thoroughly analyzing the link between cirrhosis, IR, and HD. It also emphasizes the vitality of targeted interventions. Cellular and molecular investigations into IR have revealed potential therapeutic targets for managing and preventing HD.

Hepatogenous diabetes: Knowledge, evidence, and skepticism

The diabetogenic potential of liver cirrhosis (LC) has been known for a long time, and the name "hepatogenous diabetes" (HD) was coined in 1906 to define the condition. Diabetes mellitus (DM) that develops as a consequence of LC is referred to as HD. In patients with LC, the prevalence rates of HD have been reported to vary from 21% to 57%. The pathophysiological basis of HD seems to involve insulin resistance (IR) and pancreatic β-cell dysfunction. The neurohormonal changes, endotoxemia, and chronic inflammation of LC initially create IR; however, the toxic effects eventually lead to β-cell dysfunction, which marks the transition from impaired glucose tolerance to HD. In addition, a number of factors, including sarcopenia, sarcopenic obesity, gut dysbiosis, and hyperammonemia, have recently been linked to impaired glucose metabolism in LC. DM is associated with complications and poor outcomes in patients with LC, although the individual impact of each type 2 DM and HD is unknown due to a lack of categorization of diabetes in most published research. In fact, there is much skepticism within scientific organizations over the recognition of HD as a separate disease and a consequence of LC. Currently, T2DM and HD are being treated in a similar manner although no standardized guidelines are available. The different pathophysiological basis of HD may have an impact on treatment options. This review article discusses the existence of HD as a distinct entity with high prevalence rates, a strong pathophysiological basis, clinical and therapeutic implications, as well as widespread skepticism and knowledge gaps.

Vascular syndromes in liver cirrhosis

Liver cirrhosis is associated with multiple vascular syndromes affecting almost all body systems. Many of these syndromes are directly related to impaired liver function and sometimes reversible after liver transplantation while others arise secondary to portal hypertension and ascites. Altered expression of angiogenic and vasoactive compounds (most importantly nitric oxide), endothelial dysfunction, dysregulated neurohormonal control, and systemic inflammatory state play differential roles in mediating homeostatic instability and abnormal vasogenic response. Important vascular features encountered in liver disease include portal hypertension, splanchnic overflow, abnormal angiogenesis and shunts, portopulmonary syndrome, hepatopulmonary syndrome, and systemic hyperdynamic circulation. Redistribution of effective circulatory volume deviating from vital organs and pooling in splanchnic circulation is also encountered in liver patients which may lead to devastating outcomes as hepatorenal syndrome. Etiologically, vascular syndromes are not isolated phenomena and vascular dysfunction in one system may lead to the development of another in a different system. This review focuses on understanding the pathophysiological factors underlying vascular syndromes related to chronic liver disease and the potential links among them. Many of these syndromes are associated with high mortality, thus it is crucial to look for early biomarkers for these syndromes and develop novel preventive and therapeutic strategies.

Diabetes in HFE Hemochromatosis

Diabetes in whites of European descent with hemochromatosis was first attributed to pancreatic siderosis. Later observations revealed that the pathogenesis of diabetes in HFE hemochromatosis is multifactorial and its clinical manifestations are heterogeneous. Increased type 2 diabetes risk in HFE hemochromatosis is associated with one or more factors, including abnormal iron homeostasis and iron overload, decreased insulin secretion, cirrhosis, diabetes in first-degree relatives, increased body mass index, insulin resistance, and metabolic syndrome. In p.C282Y homozygotes, serum ferritin, usually elevated at hemochromatosis diagnosis, largely reflects body iron stores but not diabetes risk. In persons with diabetes type 2 without hemochromatosis diagnoses, serum ferritin levels are higher than those of persons without diabetes, but most values are within the reference range. Phlebotomy therapy to achieve iron depletion does not improve diabetes control in all persons with HFE hemochromatosis. The prevalence of type 2 diabetes diagnosed today in whites of European descent with and without HFE hemochromatosis is similar. Routine iron phenotyping or HFE genotyping of patients with type 2 diabetes is not recommended. Herein, we review diabetes in HFE hemochromatosis and the role of iron in diabetes pathogenesis in whites of European descent with and without HFE hemochromatosis.

The Gastrointestinal Circulation: Physiology and Pathophysiology

The gastrointestinal (GI) circulation receives a large fraction of cardiac output and this increases following ingestion of a meal. While blood flow regulation is not the intense phenomenon noted in other vascular beds, the combined responses of blood flow, and capillary oxygen exchange help ensure a level of tissue oxygenation that is commensurate with organ metabolism and function. This is evidenced in the vascular responses of the stomach to increased acid production and in intestine during periods of enhanced nutrient absorption. Complimenting the metabolic vasoregulation is a strong myogenic response that contributes to basal vascular tone and to the responses elicited by changes in intravascular pressure. The GI circulation also contributes to a mucosal defense mechanism that protects against excessive damage to the epithelial lining following ingestion of toxins and/or noxious agents. Profound reductions in GI blood flow are evidenced in certain physiological (strenuous exercise) and pathological (hemorrhage) conditions, while some disease states (e.g., chronic portal hypertension) are associated with a hyperdynamic circulation. The sacrificial nature of GI blood flow is essential for ensuring adequate perfusion of vital organs during periods of whole body stress. The restoration of blood flow (reperfusion) to GI organs following ischemia elicits an exaggerated tissue injury response that reflects the potential of this organ system to generate reactive oxygen species and to mount an inflammatory response. Human and animal studies of inflammatory bowel disease have also revealed a contribution of the vasculature to the initiation and perpetuation of the tissue inflammation and associated injury response.

Splanchnic vasodilation and hyperdynamic circulatory syndrome in cirrhosis

Portal hypertension is a clinical syndrome which leads to several clinical complications, such as the formation and rupture of esophageal and/or gastric varices, ascites, hepatic encephalopathy and hepato-renal syndrome. In cirrhosis, the primary cause of the increase in portal pressure is the enhanced resistance to portal outflow. However, also an increase in splanchnic blood flow worsens and maintains portal hypertension. The vasodilatation of arterial splanchnic vessels and the opening of collateral circulation are the determinants of the increased splanchnic blood flow. Several vasoactive systems/substances, such as nitric oxide, cyclooxygenase-derivatives, carbon monoxide and endogenous cannabinoids are activated in portal hypertension and are responsible for the marked splanchnic vasodilatation. Moreover, an impaired reactivity to vasoconstrictor systems, such as the sympathetic nervous system, vasopressin, angiotensin II and endothelin-1, plays a role in this process. The opening of collateral circulation occurs through the reperfusion and dilatation of preexisting vessels, but also through the generation of new vessels. Splanchnic vasodilatation leads to the onset of the hyperdynamic circulatory syndrome, a syndrome which occurs in patients with portal hypertension and is characterized by increased cardiac output and heart rate, and decreased systemic vascular resistance with low arterial blood pressure. Understanding the pathophysiology of splanchnic vasodilatation and hyperdynamic circulatory syndrome is mandatory for the prevention and treatment of portal hypertension and its severe complications.

Overnight glucose infusion suppresses renal ammoniagenesis and reduces hyperammonaemia induced by a simulated bleed in cirrhotic patients

BACKGROUND

A simulated upper gastrointestinal (UGI) bleed in cirrhotic patients has been shown to induce hyperammonaemia. The kidney was the site of this exaggerated ammoniagenesis with alanine as substrate. Administration of alanine to decompensated cirrhotic patients did not change hepatic gluconeogenesis, but resulted in increased ammoniagenesis. We hypothesise that reduced hepatic glycogen stores result in hyperglucagonaemia which may drive increased renal gluconeogenesis and therefore alanine uptake and renal ammoniagenesis.

AIM

To determine whether an overnight glucose infusion lowers renal ammoniagenesis by reducing hyperglucagonaemia and renal ammoniagenesis.

METHODS

Patients with decompensated cirrhosis were studied in a cross-over design. An UGI bleed was simulated via intragastric administration of an amino acids mixture mimicking the haemoglobin molecule after a 12-h overnight fast (F-group) or after a 12-h treatment with 20% glucose solution (G-group).

RESULTS

Before the simulated bleed the glucagon levels were 21 (15-31) pmol/L in the F-group and 15 (9-21) pmol/L in the G-group (P < 0.01). After the simulated bleed, arterial ammonia levels increased in both groups [F-group: 73-118 μmol/L (P = 0.01); G-group 64-87 μmol/L (P = 0.01)]. The enhancement of hyperammonaemia was significantly higher in the F-group (45 [19-71] μmol/L) compared with the G-group (23 [13-39] μmol/L) (P = 0.01). The difference in renal ammoniagenesis during the simulated bleed in the F-group was 399 (260-655) nmol/kg/bwt/min and was significantly higher than in the G-group 313 (1-498) nmol/kg/bwt/min (P = 0.05).

CONCLUSIONS

Overnight glucose infusion results in reduced renal ammoniagenesis and attenuates ammonia levels. These observations have implications for the development of nutritional strategies in hyperammonaemic patients.

Body protein metabolism and plasma amino acids in cirrhosis of the liver. The effect of varying the branched chain amino acid content of intravenous amino acid solutions

Body protein metabolism and plasma amino acids were measured in 37 patients with stable cirrhosis of the liver to assess the effect of disease severity (Child's classification). Thirty two patients underwent a second series of measurements while nutrition was administered intravenously as one of five different infusions. Four infusions were formulations of amino acids with dextrose of varying branched chain amino acid content (100%, 53%, 35% and 16% branched chain to total amino acids). The fifth infusion was dextrose alone. No differences were detected in body protein synthesis and breakdown between patients on the basis of disease severity although some small differences were noted in the plasma amino acids. Infusion of dextrose alone and the 16% BCAA solution led to negative protein balance and a lowering of the plasma branched chain amino acid concentrations. Improved protein balance was observed with 35% BCAA, this solution also lowered the levels of methionine and aromatic amino acids. In those patients given 53% BCAA protein balance was achieved and the plasma branched chain amino acids elevated. Protein balance also occurred with 100% BCAA, in association with marked increases in underlying protein synthesis and breakdown and with this infusion there were marked elevations in the plasma branched chain amino acids and depression of the other plasma amino acids. The increase in protein breakdown with this last formulation was unexpected and may be harmful. On the basis of these findings it is suggested that the composition of the 53% and 35% solutions may be optimal to protein metabolism and manipulation of plasma amino acids in this group of patients.

Signals for glucagon secretion

The normal physiological role of glucagon is in controlling hepatic glucose output. Glucagon subserves the role of homeostasis by maintaining plasma glucose and of a stress hormone by producing hyperglycaemia. While control of glucagon release by circulating metabolites and also other hormones is clearly important, it seems likely that the nervous system exerts an over-riding influence. The parasympathetic nervous system maintains homeostasis and the sympathetic acts in stress. Glucagon levels are found to be high in cirrhosis and also after acute hepatic failure. It is likely that these changes in glucagon concentration are secondary to metabolic abnormalities. While some glucagon is cleared by the liver, a similar clearance is seen by many other tissues and it is not likely that the elevation of glucagon seen in liver failure is due solely to a gross deficiency of glucagon clearance. No liver abnormality is seen in the glucagonoma syndrome, where glucagon concentration are chronically high, or in patients who have had a total pancreatectomy, where plasma glucagon is undetectably low. It thus seems unlikely that liver mass is importantly controlled by glucagon.

Splanchnic and systemic vasodilation: the experimental models

Experimental models are a sine qua non condition for unraveling the specific components and mechanisms contributing to vascular dysfunction and arterial vasodilation in portal hypertension. Moreover, a careful selection of the type of animal model, vascular bed, and methodology is crucial for any investigation of this issue. In this review, some critical aspects related to experimental models in portal hypertension and the techniques applied are highlighted. In addition, a detailed summary of the mechanisms of arterial vasodilation in portal hypertension is presented. First, humoral and endothelial vasodilators, predominantly nitric oxide but also carbon monoxide and endothelium-derived hyperpolarizing factor, and others are discussed. Second, time course and potential stimuli triggering and/or perpetuating splanchnic vasodilation are delineated. Finally, a brief general overview of vascular smooth muscle signaling sets the stage for a discussion on cotransmission, receptor desensitization, and the observed impairment in vasoconstrictor-induced smooth muscle contraction in the splanchnic and systemic circulation during portal hypertension.

Vasoactive factors and hemodynamic mechanisms in the pathophysiology of portal hypertension in cirrhosis

Portal hypertension is primarily caused by the increase in resistance to portal outflow and secondly by an increase in splanchnic blood flow, which worsens and maintains the increased portal pressure. Increased portal inflow plays a role in the hyperdynamic circulatory syndrome, a characteristic feature of portal hypertensive patients. Almost all the known vasoactive systems/substances are activated in portal hypertension, but most authors stress the pathogenetic role of endothelial factors, such as COX-derivatives, nitric oxide, carbon monoxide. Endothelial dysfunction is differentially involved in different vascular beds and consists in alteration in response both to vasodilators and to vasoconstrictors. Understanding the pathogenesis of portal hypertension could be of great utility in preventing and curing the complications of portal hypertension, such as esophageal varices, hepatic encephalopathy, ascites.

The future treatment of portal hypertension

Increased understanding of the hyperdynamic circulation syndrome has resulted in novel therapeutic approaches, some of which have already reached clinical practice. Central to the hyperdynamic circulation syndrome is an imbalance between the increase in different vasodilators (foremost among which is nitric oxide) and the compensatory increase in vasoconstrictors--usually accompanied by a blunted response. This chapter discusses the role of endothelin in the pathogenesis of the syndrome and in future treatment approaches. A relatively new area of research in this field is the role of infection and inflammation in the initiation and maintenance of the hyperdynamic circulation syndrome. The use of antibiotics in the setting of acute variceal bleeding is standard practice. Studies have suggested that chronic manipulation of the intestinal flora could have beneficial effects in the treatment of portal hypertension. The bile salts are another novel and interesting target. Although their vasoactive properties have been known for some time, recent data demonstrate that their effects could be central in the pathogenesis of the hyperdynamic circulation syndrome, and that manipulation of the composition of the bile acid pool could be a therapeutic approach to portal hypertension. Finally, hypoxia and angiogenesis play a role in the development of portal hypertension and the formation of collaterals. This role needs to be further defined but it appears likely that this phenomenon is yet another target for therapeutic intervention.

Hepatitis C virus down-regulates insulin receptor substrates 1 and 2 through up-regulation of suppressor of cytokine signaling 3

The pathogenesis of hepatitis C virus (HCV)-associated insulin resistance remains unclear. Therefore, we investigated mechanisms for HCV-associated insulin resistance. Homeostasis model assessment for insulin resistance was increased in patients with HCV infection. An increase in fasting insulin levels was associated with the presence of serum HCV core, the severity of hepatic fibrosis and a decrease in expression of insulin receptor substrate (IRS) 1 and IRS2, central molecules of the insulin-signaling cascade, in patients with HCV infection. Down-regulation of IRS1 and IRS2 was also seen in HCV core-transgenic mice livers and HCV core-transfected human hepatoma cells. Carbobenzoxy-l-leucyl-l-leucyl-l-leucinal, a potent proteosomal proteolysis inhibitor, blocked down-regulation of IRS1 and IRS2 in HCV core-transfected hepatoma cells. In human hepatoma cells, HCV core up-regulated suppressor of cytokine signaling (SOCS) 3 and caused ubiquitination of IRS1 and IRS2. HCV core-induced down-regulation of IRS1 and IRS2 was not seen in SOCS3(-/-) mouse embryonic fibroblast cells. Furthermore, HCV core suppressed insulin-induced phosphorylation of p85 subunit of phosphatidylinositol 3-kinase and Akt, activation of 6-phosphofructo-2-kinase, and glucose uptake. In conclusion, HCV infection changes a subset of hepatic molecules regulating glucose metabolism. A possible mechanism is that HCV core-induced SOCS3 promotes proteosomal degradation of IRS1 and IRS2 through ubiquitination.

Plasma endothelin-1 levels in liver cirrhosis

The role of circulating endothelin- , a potent vasoconstricting peptide, in liver cirrhosis is still controversial. It has been postulated that endothelin-1 may play a role in the circulatory derangement occurring in cirrhotic subjects, and increased plasma endothelin-1 levels have been reported in these patients. In this study we looked for a relationship between the severity of the liver disease according to Child's classification and plasma endothelin-1 concentrations in a group of cirrhotic patients compared with a healthy control group. Twenty-two cirrhotic patients and 10 healthy controls, matched for sex and age, were selected for study after informed consent. The etiology of cirrhosis was posthepatitis B in 8 of 22 cases, posthepatitis C in 13 of 22 cases, and alcoholism in 1 patient. According to Child's classification, 6 patients were in class A, 6 in class B, and 10 in class C. Plasma endothelin-1 was measured by a commercial RIA kit (Amersham UK). Mean +/- SD plasma endothelin-1 levels were 8.8 +/- 0.9 pg/ml in controls and 9.2 +/- 1.1 pg/ml in all cirrhotic patients (P > 0.05). In each sub-group of cirrhotics, plasma endothelin- was 8.6 +/- 1.2 pg/ml in Child A, 8.9 +/- 1.9 pg/ml in Child B, and 10.6 +/- 1.5 pg/ml in Child C groups, respectively. There were no statistical differences between control subjects and Child A and B cirrhotic patients (P > 0.05). A significant increase in endothelinl was observed only in the Child C group versus either group A or B (P = 0.004). Our results show that alterations of circulating endothelin-1 do not occur in all cirrhotic patients; higher plasma levels than controls are only detectable in patients with more-severe hepatic failure. We do not know whether increased endothelin-1 levels are a consequence of hemodynamic disorders occurring in the advanced phase of liver cirrhosis or play a pathogenic role.

Effect of mesocaval interposition shunting and repeated sclerotherapy on blood levels of gastrointestinal regulatory peptides, amino acids, and lysosomal enzymes--a prospective randomised trial

AIMS/BACKGROUND

Patients with liver cirrhosis and portal hypertension frequently exhibit a multitude of alterations of hormones and metabolism, but the relation of these alterations to liver function, degree of blood shunting, and hepatic encephalopathy remains unclear.

METHODS

Twenty-four patients were randomised to mesocaval interposition shunt (MIS) and 21 patients to repeated sclerotherapy (ST). Several peptide hormones, amino acids and lysosomal enzymes were monitored during a 4 year follow-up period.

RESULTS

Insulin and glucagon levels were elevated in the MIS group compared to pre-therapy levels, whereas the gastrin level was significantly higher in the ST group. Pancreatic polypeptide, somatostatin and vasoactive intestinal peptide levels were not affected by either treatment. The branched chain amino acids valine, leucine and isoleucine serum levels were all elevated after ST, and the arginine, proline and tyrosine levels were higher in the MIS group at follow-up. Other amino acids were not changed, neither were the lysosomal enzymes beta-hexosaminidase nor beta-glucoronidase during this longterm follow-up.

CONCLUSION

MIS or repeated ST treatment only affected serum levels of hormones, amino acids and lysosomal enzymes to a limited extent. In this trial, the type of treatment had only a small influence on these parameters during long term follow-up.

Quantification of gluconeogenesis in cirrhosis: response to glucagon

BACKGROUND & AIMS

Accelerated starvation and early recruitment of alternate fuels in cirrhosis have been attributed to reduced availability of hepatic glycogen. The aim of this study was to measure gluconeogenesis (as a marker of protein oxidation) in relation to total glucose production and glucagon-stimulated glycogenolysis.

METHODS

Glucose and urea production, gluconeogenesis, and glycogenolysis were calculated using stable isotope methods before and during glucagon infusion (3 ng. kg-1. min-1) in 5 cirrhotic patients and 5 matched controls before and after glycogen repletion.

RESULTS

In the basal state, cirrhotic patients had a normal rate of glucose production, but the contribution of gluconeogenesis was increased (74.3% +/- 4.1% vs. 55. 6% +/- 12.1%; P < 0.005). Glycogen repletion normalized the rate of gluconeogenesis. The glycemic response to glucagon (3 ng. kg-1. min-1) was blunted in cirrhotic patients because of a lower rate of glycogenolysis (0.63 +/- 0.23 vs. 1.22 +/- 0.23 mg. kg-1. min-1; P < 0.01) and was not affected by glycogen repletion. Despite increased gluconeogenesis, the simultaneously measured rate of urea synthesis was lower in cirrhotic patients (3.11 +/- 1.02 vs. 5.0 +/- 1.0 mg/kg; P < 0.05).

CONCLUSIONS

These data show that in cirrhosis, glucose production is sustained by an increased rate of gluconeogenesis. The hepatic resistance to glucagon action is not caused by reduced glycogen stores.

Cardiovascular effects of alcohol

The ingestion of one or two alcoholic drinks can affect heart rate, blood pressure, cardiac output, myocardial contractility, and regional blood flow. These actions generally are not clinically important. In the presence of cardiovascular disease, however, even such small quantities of alcohol might result in transient unfavorable hemodynamic changes. Moreover, alcohol abuse can produce cardiac arrhythmias, hypertension, cardiomyopathy, stroke, and even sudden death. In contrast, moderate alcohol use produces changes that have an overall favorable effect on atherosclerotic-related vascular diseases. Because cardiovascular disease due to atherosclerosis is the leading cause of death in Western society, this desirable effect of alcohol use outweighs its detrimental actions, resulting in favorable findings in population studies. Nevertheless, the body of evidence argues against encouraging alcohol use for its cardiovascular effects.

Pattern of plasma cyclic nucleotides and related hormones in liver cirrhosis and hepatocellular carcinoma

To evaluate the pattern of plasma cyclic adenosine 3',5'-monophosphate, cyclic guanosine 3',5'-monophosphate, atrial natriuretic factor and glucagon levels in different stages of chronic liver diseases, we measured these variables in 20 normal subjects, 25 patients with genetic hemochromatosis, associated with liver cirrhosis in 19 cases and not in six, eight patients with compensated and 15 with decompensated alcoholic or posthepatitic cirrhosis, and 12 with hepatocellular carcinoma. All variables were within the normal range in non-cirrhotic hemochromatotic patients. Cyclic adenosine 3',5'-monophosphate levels were within the normal range (9.5-15.7 nmol/l) in hemochromatotic cirrhotics and elevated in other patients. Cyclic guanosine 3',5'-monophosphate, atrial natriuretic factor and glucagon were above the normal ranges (1.92-5.91 nmol/l, 8.8-62.7 ng/l, and 39-165 ng/l, respectively) in most patients with cirrhosis both with and without hemochromatosis and in most individuals with hepatocellular carcinoma. Cyclic guanosine 3',5'-monophosphate correlated with atrial natriuretic factor in the former groups but not in the latter. These findings indicate that glucagon and atrial natriuretic factor hypersecretion is an early event in cirrhosis, regardless of its etiology. In hepatocellular carcinoma, the underlying cirrhosis may account for most hormonal and metabolic changes although cyclic guanosine 3',5'-monophosphate increases could also be due to the neoplastic process per se.

Hyperglucagonaemia in cirrhotic patients and its relationship to the severity of cirrhosis and haemodynamic values

Plasma glucagon concentrations were measured in 160 cirrhotic patients (Pugh's grade A in 52 patients, Pugh's grade B in 64 patients and Pugh's grade C in 44 patients). These values were compared with plasma glucagon concentrations in 57 age and sex-matched healthy subjects. Systemic and portal haemodynamic measurements, effective renal plasma flow and creatinine clearance were recorded for each patient. Plasma glucagon levels were significantly increased in cirrhotic patients compared with healthy subjects. In addition, plasma glucagon levels were higher in cirrhotic patients with ascites than in those without ascites and were increased in relation to the severity of cirrhosis as assessed by Pugh's score. Multiple linear regression found that only Child-Pugh's score was estimated to be an independent predictor of hyperglucagonaemia in cirrhotic patients. However, in patients with different degrees of oesophageal varices and in patients without oesophageal varices, plasma glucagon concentrations were no different among the different groups of patients, but were still higher than plasma glucagon concentrations in healthy subjects. In contrast, plasma glucagon levels were negatively correlated with mean arterial pressure and systemic vascular resistance. The results of the present study suggest that impairment of liver function plays, in part, a role in increased plasma glucagon levels observed in patients with cirrhosis. In addition, these data support the hypothesis that hyperglucagonaemia may contribute, at least in part, to the pathogenesis of peripheral arterial vasodilatation in cirrhosis with portal hypertension.

Necrolytic acral erythema: a cutaneous marker of viral hepatitis C

BACKGROUND

Necrolytic acral erythema (NAE) is a distinctive skin lesion that was found to affect the dorsa of the feet of seven patients having active viral hepatitis C. Necrolytic acral erythema occurs in the form of well circumscribed dusky erythematous areas that develop flaccid blisters in their early stages and a hyperkeratotic surface in their chronic form. Microscopically, lesions of NAE are similar to those of other necrolytic erythemas such as necrolytic migratory erythema, pellagra, and zinc deficiency.

METHOD

Seven patients with NAE were included in this study. These patients underwent microscopic examination of punch biopsy specimens of the affected skin, abdominal sonography, CT scan of pancreas, and a liver biopsy. Blood samples were obtained for complete blood picture, serum glucose, zinc, amino acids, liver function tests, and markers of hepatitis.

RESULTS

All patients with NAE were found to have hepatitis C by ELISA and PCR.

CONCLUSIONS

Necrolytic acral erythema is a distinctive type of necrolytic erythemas that was observed to occur almost exclusively with viral hepatitis C. Therefore, it should be considered an important cutaneous marker of hepatitis C, particularly in areas showing a high incidence of this form of hepatitis.

Mediators, cytokines, and growth factors in liver-lung interactions

Multiple mediators have been implicated in the interactions between the liver and the lungs in various disease states. The best characterized mediator of liver-lung interaction is alpha 1-antitrypsin. Several cytokines and mediators may be involved in the pathogenesis of the hepatopulmonary syndrome and in the cytokine cascades that are activated in systemic inflammatory states such as acute respiratory distress syndrome. Hepatocyte growth factor or scatter factor is a recently described peptide with a broad range of biologic effects that may mediate lung-liver interactions.

Decreased vascular reactivity of portal vein in rats with portal hypertension

BACKGROUND/AIMS

Vascular hyporesponsiveness in portal hypertension is well documented in the arterial tissue. The present study aimed to investigate the possible changes in the portal vein from portal hypertensive rats.

METHODS

Portal hypertension was induced by partial portal vein ligation. Fourteen days after surgery, portal veins were removed for contractile study and measurement of cAMP, cGMP and [Ca2+]i.

RESULTS

In vitro tension preparation showed a decreased maximum response to norepinephrine in portal vein of portal vein ligated rats (38.3 +/- 4.1 vs. 23.4 +/- 1.5 mN/mm2). The pA2 values of WB4101 and yohimbine (alpha1- and alpha2-adrenoceptor antagonist, respectively) were not different between the two groups. Tissue cAMP (14.4 +/- 0.9 vs. 12.2 +/- 0.7 pmol/mg protein), but not cGMP, content was increased and intracellular calcium [Ca2+]i levels (247 +/- 9 vs. 281 +/- 13 nM) were decreased in portal vein ligated rats.

CONCLUSIONS

These results showed that in portal vein from portal vein ligated rats, vascular hyporesponsiveness and an increase in basal cAMP content and a decrease in basal [Ca2+]i were observed.

Glucagon, stress, and portal hypertension. Plasma glucagon levels and portal hypertension in relation to anesthesia and surgical stress

Glucagon has been proposed as the mediator of splanchnic hyperemia in portal hypertension. Employing an assay specific for pancreatic glucagon, we reevaluated the relationship between this peptide and portal hypertension in the portal vein (PV)-stenosed rat model addressing, in particular, the effects of anesthesia and surgical stress. Plasma glucagon levels were similar in sham-operated and portal hypertensive rats: glucagon, sham vs PV stenosed: 110.7 +/- 17.1 pmol/liter vs 140.6 +/- 23.3 pmol/liter (NS). Furthermore, plasma levels of glucagon and the related peptide VIP were not significantly influenced by anesthesia or surgical stress, and levels remained similar under all conditions in sham-operated and PV-stenosed animals. We conclude that pancreatic glucagon is not elevated in the PV-stenosed rat; differences between these results and those describing hyperglucagonemia in this model cannot be explained on the basis of a differential response to stress but may reflect differences in glucagon assay system.

Vasodilatory state of decompensated cirrhosis: relation to hepatic dysfunction, ascites, and vasoactive substances

The objective of this study was to determine the relations between the hallmark circulatory finding of decompensated cirrhosis, a reduced systemic vascular resistance (SVR), and the indices of hepatic decompensation, the accumulation of ascites, and the concentrations of various vasoactive substances. At a university-affiliated teaching hospital, eighteen hospitalized patients with cirrhosis and 18 age- and sex-matched healthy subjects were used. This was a case-control study. Measurements included cardiac dimensions and indices derived from echocardiograms and Doppler studies, abdominal ultrasound estimates of ascites, indices of hepatic function, and various serum (S) and urinary (U) substances. Results showed that cirrhotics had increased left atrial and left ventricular dimensions, left ventricular mass, heart rate, cardiac output (CO), transvalvular velocities, and a decreased SVR. SVR was related to hepatic dysfunction, as reflected by an abnormal prothrombin time ratio (r = -0.64, p = 0.006), and also related to overall severity of liver disease as estimated by the Child-Pugh score (r = -0.53, p = 0.044). Although cirrhotics with ascites generally had a reduced SVR, estimates of ascites were directly related to SVR (r = 0.57, p = 0.03) and inversely related to CO (r = -0.53, p = 0.04). Concentrations of S and U digoxin-like immunoreactive substance (DLIS) were also increased, but the concentrations of S glucagon and estradiol were not elevated. The accumulations of S and U DLIS, S glucagon, and S estradiol were all related to hepatic dysfunction.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of portal vein stenosis and superior mesenteric vein ligation on mesenteric venous pressure and porta-systemic shunting in the rat

We studied the relative contributions of splanchnic congestion and porta-systemic shunting to the maintenance of experimental portal venous hypertension. Three groups of rats were prepared: portal vein-stenosed, superior mesenteric vein-ligated and sham operated. Though elevated in both operated groups compared to controls, mesenteric venous pressure was highest in the portal vein-stenosed animals (PV vs SMV vs Sham: 19.6 +/- 1.3 vs 15.6 +/- 0.7 vs 13 +/- 0.6; p < .05 PV and SMV vs Sham, and PV vs SMV) despite the presence of 50% porta-systemic shunting in the portal vein-stenosed animals. Shunting was negligible in the other two groups. Peripheral plasma glucagon and vasoactive intestinal peptide (VIP) levels were similar in all three groups. We conclude that mesenteric congestion alone plays a minor role in the pathogenesis of portal hypertension, which may instead be related to the porta-systemic shunting of vasoactive substances other than glucagon and VIP.

Impaired function of pancreatic islets from rats with portal hypertension resulting from cirrhosis and partial portal vein ligation

Increased circulating insulin and glucagon levels are a common observation in patients with cirrhosis, as well as in portal hypertensive models. Hyperinsulinemia and hyperglucagonemia may be caused either by increased beta- and alpha-cell secretion or by defective hepatic clearance of these hormones. To elucidate whether an abnormal endocrine pancreatic function might contribute to the hyperinsulinism or to the hyperglucagonism observed in chronic portal hypertension, insulin and glucagon secretion were measured in vitro in isolated pancreatic islets from rats with partial portal vein ligated and rats with cirrhosis caused by carbon tetrachloride poisoning. Both rats with partial portal vein ligation and rats with cirrhosis caused by carbon tetrachloride poisoning exhibited hyperinsulinism and hyperglucagonism as compared with control rats. Isolated pancreatic islets from both experimental portal hypertensive models showed an impaired insulin secretion after glucose stimulation. On the contrary, glucagon secretion was significantly increased, and there was a markedly enhanced response to arginine. This increased in vitro glucagon production could not be corrected, even in the presence of high glucose concentrations in the incubation medium. Therefore our data show that although hyperglucagonism in rats with partial portal vein ligation and in rats with cirrhosis caused by carbon tetrachloride poisoning is promoted by an enhanced alpha-cell secretion, hyperinsulinism is associated with impaired beta-cell secretion.

Rationale and efficacy of specialized enteral nutrition

Enteral nutrition, whenever feasible, is the preferred route of nutrition support. There has been a tremendous increase in the number of enteral products available on the market. Many of these are designed for use in specific disease states. This review will summarize the scientific rationale and supporting research for specialized enteral nutrition.

Failure of glucagon to stimulate hepatic glycogenolysis in well-nourished patients with mild cirrhosis

The ability of glucagon to stimulate hepatic glucose production (HGP) was studied in clinically stable cirrhotic patients (n = 8) who had, based on long-term follow-up evaluation, relatively good liver function (Child-Pugh A) and whose dietary intake and physical characteristics were comparable to those of healthy control subjects (n = 8). Plasma glucagon concentration was slightly but not significantly increased in cirrhotic patients versus control subjects in the basal state (190 +/- 41 v 126 +/- 24 pg/mL, P = NS) and during a continuous 180-minute glucagon infusion at 3 ng/kg/min (349 +/- 56 v 243 +/- 37, P = NS). The increment in plasma glucagon level (+164 +/- 57 v +127 +/- 35, P = NS) also was slightly greater in the cirrhotic group. HGP (measured with [6-3H]-glucose) in the basal state was similar in cirrhotic and control subjects (1.79 +/- 0.09 v 1.94 +/- 0.15 mg/kg/min, P = NS). In cirrhotic patients, stimulation of HGP by glucagon was blunted during the first 15 to 30 minutes of the infusion period (representing glucagon's predominant effect on glycogenolysis; 0.23 +/- 0.20 v 1.06 +/- 0.19 mg/kg/min, P < .05), but it was not different from that in control subjects during the remaining course of the experiment (30 to 180 minutes). Basal plasma insulin and C-peptide concentrations did not change from baseline during the glucagon infusion in cirrhotics, whereas they increased slightly but not significantly in controls. These data demonstrate that even in the early stages of cirrhosis, the liver is resistant to the stimulatory effect of glucagon on hepatic glycogenolysis.(ABSTRACT TRUNCATED AT 250 WORDS)

Hepatic regulation of pancreatic alpha-cell function

The direct feedback regulation between the endocrine gland and its target organ is an expected biological relationship. However, such a phenomenon is far from being well established in the case of the endocrine pancreas and its major target organ, the liver, especially since plasma glucose has been established as the prime regulator. In this perspective, I have reexamined the feedback regulation between plasma glucose and glucagon secretion by the pancreatic alpha cell. Surprisingly, available data in the literature appear to document a frequent breakdown of this well-established interdependence between plasma glucose and pancreatic alpha cells, as reflected by a sustained elevation of plasma glucagon levels in several physiologic and pathologic states with concurrent euglycemia or hyperglycemia. Moreover, normal or low glucagon concentrations in the presence of fasting hypoglycemia in patients with insulinoma or non-islet cell tumors secreting insulin-like peptides and in patients with hepatic glycogen storage disorders may enhance our hypothesis that plasma glucose level may not be the major regulator of glucagon secretion. Extensive data in the literature show that hyperglucagonemic states are characterized by a unique metabolic environment, namely hepatic glycogen depletion. Similarly, hepatic glycogen stores are abundant in the presence of normal or low glucagon concentrations. These findings imply a distinct relationship between hepatic glycogen content and plasma glucagon level.(ABSTRACT TRUNCATED AT 250 WORDS)

Sulfur amino acid metabolism in hepatobiliary disorders

Sulfur amino acid metabolism was studied in patients with mild to severe forms of liver dysfunction and compared with that of healthy controls. Patients with mild liver dysfunction (for example, Gilbert's syndrome) had a normal sulfur amino acid metabolism. With increased inflammatory activity and cirrhosis (for example, chronic active hepatitis, alcohol-induced cirrhosis, and hepatic coma) a decreased ability to metabolize methionine (to cysteine, with cystathionine accumulation) and cysteine (to inorganic sulfate, with thiosulfate and N-acetylcysteine accumulation) was found. In contrast, transaminative metabolism of sulfur amino acids was preserved in patients with advanced forms of liver dysfunction, suggesting that transamination of sulfur amino acids is performed not only in the liver but also in extrahepatic tissues. Some implications of these findings are discussed.

Glucagon-independent renal hyperaemia and hyperfiltration after an oral protein load in Child A liver cirrhosis

The work was designed to study the effects of a meat meal on glomerular filtration rate (GFR), renal plasma flow (RPF), and plasma concentrations of glucagon, insulin, growth hormone, renin, aldosterone, total amino acids, and NH3 in healthy humans (H) as well as in patients with Child A liver cirrhosis (LC). The meat meal produced renal hyperaemia and hyperfiltration without changes in the filtration fraction. Fractional Na excretion in urine increased significantly after the meat meal only in LC. Hyperinsulinaemia and hyperglucagonaemia were seen at baseline in LC and were not affected by the meat meal, whereas in H glucagon concentration increased significantly over baseline within 30 min from the meat meal and insulin within 60 min. Growth hormone concentration was normal at baseline in LC and increased significantly 120-180 min after the meal, whereas it was not affected in H. Renin and aldosterone were stable in both H and LC. Plasma amino acid concentration began to increase 60 min after the meat meal, when hyperfiltration was present. The data indicate that in human Child A cirrhosis of the liver renal haemodynamic response to a meat meal is independent of changes in glucagon.

Circulatory changes induced by portal venous diversion and mesenteric hypertension in rats

We studied the hemodynamics in four groups of rats with combinations of mesenteric hypertension and portal diversion. Operations created three groups with mesenteric hypertension and different degrees of portal venous diversion: mesenteric vein stenosis, portal vein stenosis and end-to-side portacaval anastomosis with mesenteric vein stenosis, the fourth group had only portacaval anastomosis. A control group had sham operations. Cardiac output, splanchnic blood flows and portosystemic shunt indices were measured with radioactive microspheres. Mesenteric venous pressures in the mesenteric-stenosed, portal-stenosed, portacaval-shunted and end-to-side portacaval anastomosis with mesenteric vein stenosis rats were, respectively, 13.5 +/- 0.6, 15.3 +/- 0.7, 4.3 +/- 0.5 and 13.0 +/- 0.9 mm Hg, which were all significantly different from controls: 8.3 +/- 0.3 mm Hg. Portosystemic shunt indices were also significantly different from each other: controls, 0.4% +/- 0.02%; mesenteric-stenosed, 5.9% +/- 2.3%; and portal-stenosed, 52.1% +/- 4.9%. Cardiac output and splanchnic visceral blood flows were significantly increased in the portal-stenosed rats and the two groups with portacaval anastomoses, with the latter two groups having the highest values. The addition of mesenteric stenosis did not change the blood flows because mesenteric-stenosed rats did not differ from controls and end-to-side portacaval anastomosis with mesenteric vein stenosis rats did not differ from rats with portacaval anastomosis alone. These results suggest that mesenteric venous hypertension per se does not affect hemodynamics but that diversion of portal venous blood from the liver is a critical factor in the development of hyperkinetic circulation in portal hypertension.

Portal hypertension--25 years of progress

Images

Role of bile acids in splanchnic hemodynamic response to chronic portal hypertension

Previous studies from our laboratory suggest that humoral factors, namely glucagon, can account for approximately 30% of the splanchnic vasodilation in rats with prehepatic portal hypertension. A reduced vascular sensitivity to norepinephrine, vasopressin, and angiotensin II may contribute to the splanchnic vasodilation. However, neither glucagon nor an altered vasoconstrictor sensitivity can fully account for the splanchnic vasodilation observed in portal hypertensive subjects. Therefore, the present study was designed to examine the role of bile acids in the splanchnic hyperemia of portal hypertension since (1) serum bile acids are elevated in portal hypertensive subjects and (2) bile acids are potent intestinal vasodilators. Prehepatic portal hypertension was induced in Sprague-Dawley rats by surgical constriction of the portal vein. Ten to 14 days after the induction of portal hypertension, the enterohepatic circulation of control and portal hypertensive rats was surgically interrupted. The animals were placed in Bollman restraint cages and allowed to recover. Eighteen to 24 hr later, the rats were anesthetized with sodium pentobarbital and regional blood flow measured with radiolabeled microspheres. Normal and portal hypertensive animals without bile fistula served as controls. Plasma bile acid levels measured by radioimmunoassay were approximately 3.8 times higher in portal hypertensive animals than in control. Bile duct cannulation effectively depleted both normal and portal hypertensive animals of their circulating bile acid pool and significantly reduced portal venous inflow in portal hypertensive but not in control rats. A role for bile acids as partial mediators of the splanchnic hyperemia of portal hypertension is suggested since bile acid depletion did not completely abolish the gastrointestinal hyperemia.

Effect of physiologic hyperinsulinemia on glucose and lipid metabolism in cirrhosis

Insulin secretion and insulin sensitivity were evaluated in eight clinically stable cirrhotic patients and in 12 controls. OGTT was normal in cirrhotics but plasma insulin response was increased approximately twofold compared with controls. Subjects received a three-step (0.1, 0.5, 1.0 mU/kg.min) euglycemic insulin clamp with indirect calorimetry, [6-3H]-glucose, and [1-14C]-palmitate. During the two highest insulin infusion steps glucose uptake was impaired (3.33 +/- 0.31 vs. 5.06 +/- 0.40 mg/kg.min, P less than 0.01, and 6.09 +/- 0.50 vs. 7.95 +/- 0.52 mg/kg.min, P less than 0.01). Stimulation of glucose oxidation by insulin was normal; in contrast, nonoxidative glucose disposal (i.e., glycogen synthesis) was markedly reduced. Fasting (r = -0.553, P less than 0.01) and glucose-stimulated (r = -0.592, P less than 0.01) plasma insulin concentration correlated inversely with the severity of insulin resistance. Basal hepatic glucose production was normal in cirrhotics and suppressed normally with insulin. In postabsorptive state, plasma FFA conc (933 +/- 42 vs. 711 +/- 44 mumol/liter, P less than 0.01) and FFA turnover (9.08 +/- 1.20 vs. 6.03 +/- 0.53 mumol/kg.min, P less than 0.01) were elevated in cirrhotics despite basal hyperinsulinemia; basal FFA oxidation was similar in cirrhotic and control subjects. With low-dose insulin infusion, plasma FFA oxidation and turnover failed to suppress normally in cirrhotics. During the two higher insulin infusion steps, all parameters of FFA metabolism suppressed normally. In summary, stable cirrhotic patients with normal glucose tolerance exhibit marked insulin resistance secondary to the impaired nonoxidative glucose disposal. Our results suggest that chronic hyperinsulinism may be responsible for the insulin resistance observed in cirrhosis.

Current concepts in the pathophysiology and treatment of portal hypertension and variceal hemorrhage

Recent studies have demonstrated that increased resistance to portal inflow is not solely responsible for the development of portal hypertension. Increased splanchnic flow has been attributed to a combination of factors, including elevated circulating levels of vasodilators and diminished sensitivity of the splanchnic vasculature to endogenous vasoconstrictors. In selected animal models of portal hypertension, increased splanchnic flow accounts for approximately 40% of the observed elevations in total portal venous pressure. Improved understanding of the pathophysiologic factors responsible for the development of portal hypertension has led to pharmacologic efforts to decrease portal pressure. Current limitations include lack of drug selectivity and specificity and inability to predict and monitor patient responses. Primary treatment options include selective portosystemic shunts, endoscopic sclerotherapy (ES), and orthotopic liver transplantation. ES is more effective in preventing recurrent variceal hemorrhage than medical treatment but is less effective than shunt surgery. In selected studies, ES better maintains hepatic function and may prolong survival compared to primary shunt surgery. ES failures occur in nearly 33% of patients, but "salvage shunts" in these patients appear to be reasonably safe and quite effective in preventing recurrent hemorrhage. Selective shunts are favored because they appear to confer a better quality of life (but not improved longevity) than conventional shunts. Liver transplantation is preferred for patients with end-stage liver disease in whom the predicted mortality of conventional surgery outweighs the survival benefit.

Transhepatic sampling during experimental porcine liver autotransplantation--its application to measurements of insulin, glucagon, and glucose

Controversy still exists in the published literature about the need for administration of intravenous glucose during liver transplantation. The ability of the grafted liver to metabolize insulin and glucagon and the appropriateness of secretion of these hormones are addressed in the present study. Two groups of pigs received unstored liver autografts, one with free infusion of 10% glucose and the other with limited infusion of 2.5% glucose solution, while attempting to maintain plasma glucose levels less than 200 mg/100 ml. In these animals, irrespective of moderate or major hyperglycemia, serum insulin levels were appropriate for blood glucose concentrations. However, in both groups, plasma glucagon levels rose three- to fourfold more than preoperative values and were inappropriate. Although facilities for measurement of blood flow were not available, application of the technique of transhepatic sampling has revealed that hepatic handling of insulin seems to be unimpaired after autograft with limited ischemia. Pancreatic secretion of glucagon, however, appeared to increase during the period immediately after revascularization. It is suggested that transhepatic sampling methods may be used in experimental transplantation to elucidate the effects of storage for prolonged periods.

Basal energy production rate and substrate use in stable cirrhotic patients

The basal energy production rate was measured using indirect calorimetry in 25 stable cirrhotic patients and 10 controls of comparable age. The endogenous substrate oxidation was also calculated by measuring urinary nitrogen excretion. The energy production rate was similar in cirrhotic patients and controls. The origins of liver disease and the degree of liver damage did not seem to influence the energy production rate. On the other hand, in cirrhotic patients, as in controls, a significant correlation was present between the energy production rate and parameters of body size, such as body weight and fat-free mass. As a consequence, cirrhotic patients with poor nutritional status, with a reduced fat-free mass, showed a lower energy production rate. The measured energy production rate was compared with the resting energy expenditure estimated by formulas commonly used in healthy individuals. The good agreement found between the measured energy production rate and calculated energy expenditure suggests that these formulas may be applied to stable cirrhotic patients in clinical practice. In cirrhotic patients, the oxidation of endogenous fat is the main contributor to basal energy production rate. The fat oxidation rate does not appear to be influenced by the hormonal pattern found in the cirrhotic patients. However, a significant correlation was present between fat oxidation and plasma free fatty acid levels. This confirms that the prevalent fat use in cirrhotic patients is supported by the greater availability of fat-derived substrates.

The effect of mesenteric venous hypertension on gut motility and absorption

The effects of portal hypertension on gut function may be mediated by venous congestion and altered circulating levels of enteric hormones and neuropeptides. We designed this study to determine the effects of chronic intestinal venous hypertension (VHT), in isolation, on gut motility and absorption. In 10 dogs, a 20- to 25-cm loop of jejunum was isolated from the fecal stream, but myoneural continuity was maintained with the proximal bowel by a seromuscular bridge. In 5 dogs, VHT was created in the loop by a fixed stenosis of its venous drainage; a sham procedure was performed in a further 5 animals. Serosal monopolar electrodes were placed in all animals. Absorptive function and myoelectrical activity were studied over a 4-week period. Venous hypertension was achieved and sustained in the VHT animals; loop vein pressures for VHT vs control in cm H2O (means +/- SEM) are: initial--29.8 +/- 1.8 vs 7.5 +/- 0.4 (P less than 0.01), and at 4 weeks--17.6 +/- 6.88 vs 7.3 +/- 0.2 (P less than 0.01). Absorption of Na+, Cl-, glucose, and water was impaired in VHT loops. Normal patterns of fasting and postprandial myoelectrical activity were preserved in the VHT animals. We conclude that chronic VHT, in the absence of portosystemic shunting, results in impaired absorption of water, glucose, and electrolytes without any change in intestinal motility.

Hemodynamic effects of glucagon in portal hypertension

It has been suggested that glucagon contributes to the pathogenesis of portal hypertension by increasing portal blood flow. This study examined this issue by assessing the hemodynamic effects of a pharmacological dose of glucagon (1 mg, intravenously) in patients with cirrhosis and portal hypertension (n = 10) and in subjects without significant liver disease (controls = n = 5). Patients with cirrhosis had much higher glucagon levels than control subjects (875 +/- 167 vs. 186 +/- 25 pg/ml, p less than 0.01) and showed blunted hemodynamic responses after glucagon administration. This occurred despite greater circulating glucagon levels, probably because of a significant prolongation of the plasma half-life of exogenously administered glucagon (4.9 +/- 0.4 vs. 2.7 +/- 0.1 min, p less than 0.1). Control subjects had marked increases in heart rate (+ 19% +/- 4%, p less than 0.01), cardiac index (+ 16% +/- 4%, p = 0.01) and arterial pressure (+ 10% +/- 3%, p less than 0.05), but corresponding changes in patients with cirrhosis (+ 7% +/- 1%, + 6% +/- 1%, and + 6% +/- 2%, respectively) were significantly less pronounced (p = 0.05), and there was a negative correlation between basal glucagon levels and the response of heart rate to glucagon injection (r = -0.804, p less than 0.001). Resistance to the systemic effects of glucagon in cirrhosis may thus be caused by a down-regulation of vascular glucagon receptors. In addition, glucagon administration caused a significant increase in portal pressure (from 18.1 +/- 1.1 to 19.0 +/- 1.2 mm Hg, p less than 0.01), as well as in azygos blood flow (from 0.54 +/- 0.03 to 0.64 +/- 0.04 L/min, + 19% +/- 4%, p less than 0.02), reflecting increased portocollateral blood flow. These findings are consistent with the hypothesis that glucagon is one of the factors contributing to the splanchnic vasodilatation and increased portal pressure of cirrhosis.