Cirrhosis, cyanosis and polycythaemia: multiple pulmonary arteriovenous anastomoses

Hepatopulmonary syndrome: role of nitric oxide and clinical aspects

Hepatopulmonary syndrome is defined by oxygenation impairment due to abnormal intrapulmonary vascular dilatations in patients with liver disease. The implication of enhanced pulmonary production of nitric oxide in the pathogenesis of intrapulmonary vascular dilatations has been demonstrated both in murine models and in human hepatopulmonary syndrome. The diagnosis of hepatopulmonary syndrome in chronic liver disease is of paramount importance, considering the fact that severe hypoxemia related to hepatopulmonary syndrome may occur in patients with well compensated liver disease and that survival is reduced in patients with hepatopulmonary syndrome relative to non hepatopulmonary syndrome patients. Priority for liver transplantation, which is presently the only cure, has been recently increased in patients with advanced hepatopulmonary syndrome.

Clinical factors that affect blood gases in non-smoking women with chronic liver disease

Chronic liver disease is often accompanied by hypoxaemia. We investigated the clinical factors that were related to the arterial oxygen tension (PaO2) in 40 women, all non-smokers with chronic liver disease. They were positive for hepatitis C virus (HCV) antibody and had no evidence of cardiopulmonary disease. Arterial blood was collected from patients at rest (> 15 min) for analysis of blood gases. We determined the correlation between blood gas tension and the clinical variables, i.e. the presence or absence of skin manifestations such as cutaneous spider nevi and palmar erythema, the presence or absence of splenomegaly, vital capacity, forced expiratory volume in one second, V25/body height, serum alanine aminotransferase (AST), serum asparate aminotransferase (ALT), serum cholinesterase, serum gamma-globulin/total protein, excretion of indocyanine green at 15 min (15-min retention rate, ICG level), blood level of ammonia, blood level of endotoxin, plasma level of glucagon and the serum level of type IV collagen-7S. The mean level of PaO2 was 78 +/- 11 (range: 43-95) torr. The mean alveolar-arterial oxygen tension gradient (A-aDO2) was 19 +/- 13 (range: 2-60) torr. Multiple regression analysis used PaO2 and A-aDO2 as objective variables, and the clinical findings as explanatory variables. The explanatory variables that were significantly correlated with blood gas values were ICG level, blood level of endotoxin and presence of skin manifestations. The ICG level showed a high correlation with blood gas values; the ICG level increased, the PaO2 decreased (r = -0.69), while the A-aDO2 showed a high positive correlation (r = +0.78, P < 0.001). Findings suggest that a reduction in hepatic blood flow and hepatocellular function interfere with the inactivation of vasoactive substances such as endotoxin by the liver, leading to the development of skin manifestations, the dilatation of intrapulmonary capillaries and the induction of hypoxaemia.

Therapeutic effects of diuretics and paracentesis on lung function in patients with non-alcoholic cirrhosis and tense ascites

BACKGROUND/AIMS

Ascites may cause or aggravate pulmonary dysfunction in patients with liver cirrhosis. Diuretics and paracentesis are the main therapies for ascites. The aim of the present study was to evaluate and compare the therapeutic effects of diuretics and large-volume paracentesis on lung function in 26 male patients with non-alcoholic cirrhosis and tense ascites.

METHODS

The patients were divided into two groups. Group A was composed of 13 subjects who were treated with diuretics including spironolactone (100-400 mg/day) and furosemide (80-320 mg/day). In group B, 13 subjects received large-volume paracentesis plus intravenous albumin (6-8 g/l ascites removed). Pulmonary function tests including spirometry, plethysmography, single-breath carbon-monoxide diffusing capacity (DLco) and arterial blood gases, were done 1 day before diuretic treatment and 1 day after termination of the study in group A patients, and 1 day before and after large-volume paracentesis in group B subjects.

RESULTS

Before treatment, the clinical and laboratory data were comparable between the two groups. After treatment, ventilatory function as evidenced by forced expiratory volume in 1 s, forced vital capacity, total lung capacity, functional residual capacity and expiratory reserve volume, and DLco increased significantly in both groups. Arterial PO2 and PCO2 increased significantly and AaPO2 (alveolar-arterial PO2 difference) decreased significantly in the subjects treated with diuretics. Nevertheless, paracentesis did not improve arterial blood gases. The changes in lung volumes, DLco and PaO2 after treatment (the data after minus those before treatment) were comparable, except that a significant decrease in AaPO2 was observed in the diuretic group.

CONCLUSIONS

Both diuretic therapy and large-volume paracentesis significantly improved the ventilatory function in patients with tense cirrhotic ascites. In terms of oxygenation improvement as evaluated by AaPO2, diuretic treatment may be superior to large-volume paracentesis.

The hepatopulmonary syndrome. Effect of liver transplantation

The role of liver transplantation as a treatment for the hepatopulmonary syndrome (HPS) has had an evolving and controversial history. Although early experience was disappointing, more recent experience has documented resolution of HPS-associated hypoxemia after liver transplantation. This article reviews the history of liver transplantation for patients with the hepatopulmonary syndrome. In addition, we discuss the clinical features that have been considered to predict successful reversibility and the time frame over which reversal occurs. Despite this evolution of thought, many basic questions still remain.

Effect of large-volume paracentesis on pulmonary function in patients with cirrhosis and tense ascites

The effect of large-volume paracentesis on lung function was evaluated in 12 male patients with cirrhosis. All underwent pulmonary function tests including spirometry, plethysmography and single-breath carbon-monoxide diffusing capacity 1 day before and after paracentesis. The amount of ascitic fluid removed ranged from 3.6 to 131 (mean +/- SD, 7.4 +/- 3.01). After paracentesis, forced vital capacity, forced expiratory volume at 1 s, total lung capacity, functional residual capacity, inspiratory capacity, expiratory reserve volume, diffusing capacity and alveolar volume increased significantly. In contrast, Kco (diffusing capacity corrected by alveolar volume) decreased significantly. After paracentesis, the increase in diffusing capacity was highly correlated with lung volumes and the amount of removed ascitic fluid. Nevertheless, a significantly negative correlation was found between the change of Kco before and after paracentesis and that of lung volumes. The increase in lung volumes and ventilation to the lower lungs with unfavorable ventilation-perfusion matching might explain the discrepancy between changes in diffusing capacity and Kco after large-volume paracentesis. In conclusion, these results suggest that pulmonary function in patients with cirrhosis and tense ascites is partly improved by large-volume paracentesis. Large-volume paracentesis might be useful for symptomatic relief in selected patients with tense ascites.

Hepatopulmonary syndrome. Clinical observations and lack of therapeutic response to somatostatin analogue

We retrospectively studied 22 patients with hepatopulmonary syndrome (HPS) evaluated at the Mayo Medical Center from 1984 to 1991. All patients had hepatic cirrhosis with clinical evidence of portal hypertension; 13 (59 percent) had severe hypoxemia while breathing room air in the supine position (PaO2 < 60 mm Hg), and 14 of 16 (88 percent) had orthodeoxia breathing room air. On the basis of angiographic observations, we defined type 1 and type 2 patterns of pulmonary vascular abnormalities in HPS. Response to 100 percent oxygen and therapeutic regimens may differ in the angiographic patterns. Substantial deterioration in PaO2 associated with clinically stable hepatic dysfunction was documented in five of seven patients studied with sequential arterial blood gas testing; four subsequently died within 48 months. Overall mortality was 41 percent, occurring a mean of 2.5 years after diagnosis. In 7 of the 22 patients, we prospectively studied the effect of somatostatin analogue given subcutaneously for 4 consecutive days. No significant improvement in PaO2 was documented while breathing room air or 100 percent oxygen (p < 0.05). We conclude that in selected patients with clinically stable hepatic dysfunction and deteriorating oxygenation, the prognosis is poor. Our data in combination with recent surgical reports suggest that liver transplantation may be the treatment of choice in patients with HPS and worsening oxygenation.

Plasma noradrenaline in cirrhosis: a study of kinetics and temporal relationship to ascites formation

The kinetics of plasma noradrenaline (NA) were studied in 14 patients with cirrhosis and ascites and 13 normal subjects. [3H]noradrenaline ([3H] NA) was infused intravenously to steady state and the spillover of NA into plasma and its clearance from plasma calculated. The increase in plasma NA in the cirrhotic patients was due to an increase in NA spillover (14.5 vs 3.9 nmol min-1m-2; P less than 0.001). NA plasma clearance was also increased in the cirrhotic patients (3.5 vs 2.11 min-1m-2; P less than 0.01). Plasma NA and dihydroxyphenylglycol (DHPG), a metabolite of NA of which a portion is formed after re-uptake of NA into sympathetic nerve endings, were then measured in 23 patients with cirrhosis and ascites, 17 patients with cirrhosis who had never had ascites, and 34 normal subjects. Both plasma NA and DHPG were significantly increased in the patients with ascites (NA 4.7, DHPG 14.7 nmol l-1 and in the patients with cirrhosis but no ascites (NA 3.8, DHPG 12.0 nmol l-1) compared with normal subjects (NA 1.9, DHPG 8.8 nmol 1-1). Therefore, the increase in plasma NA in cirrhosis is due to increased activity of the sympathetic nervous system rather than interference with the metabolism of NA or impaired neuronal uptake of NA. This increase appears to precede the development of ascites.

Reduction of intrapulmonary shunt and resolution of digital clubbing associated with primary biliary cirrhosis after liver transplantation

This report describes a patient with marked hypoxemia caused by intrapulmonary shunt associated with primary biliary cirrhosis. Liver transplantation resulted in resolution of digital clubbing and reduction of intrapulmonary shunt as demonstrated by normalization of room air arterial blood gases, reduction in shunt fraction and normalization of the indocyanine-enhanced echocardiogram and perfusion lung scan. This patient's course challenges the conventional notion that intrapulmonary shunting associated with chronic liver disease does not reverse after liver transplantation.

Severe hypoxemia and liver disease

Severe hypoxemia and orthodeoxia in patients with chronic liver disease is uncommon, but, when present, it is incapacitating. The purpose of this study was to determine the distribution of alveolar ventilation-perfusion (VA/Q) in six patients with mild liver disease and severe hypoxemia (PaO2 at rest in sitting or standing position ranged from 35 to 67 mm Hg). Orthodeoxia was documented with improvement in PaO2 in the supine position in each patient (PaO2 at rest in supine position ranged from 46 to 75 mm Hg). VA/Q distribution was measured by the multiple inert gas elimination technique. The dispersion of VA/Q was increased with small portions of the cardiac output (0.5 to 14.8%) perfusing low VA/Q areas (O less than VA/Q less than 0.1). Another major finding was a large right-to-left shunt (VA/Q less than 0.005) that ranged from 4 to 28%. The VA/Q mismatching and the right-to-left shunt both contributed to the hypoxemia. The predicted PaO2 was 5.5 mm Hg (p less than 0.01) larger than the measured PaO2. In each patient, the mean pulmonary artery pressure was low and the cardiac output was elevated. These results show that the low PaO2 in these patients was due to both increased right-to-left shunt and VA/Q mismatching, but impaired diffusion could not be ruled out.

Ventricular function in cirrhosis and portasystemic shunt: a two-dimensional echocardiographic study

Cardiovascular alterations such as increased heart rate, high cardiac output, reduced systemic vascular resistance, and in most of the cases, increased contractility parameters have been recognized in patients with advanced liver disease. Some investigators define a cirrhotic cardiomyopathy as a hyperdynamic failure of the heart. Consequently, in patients with cirrhosis, the risk of developing further circulatory deterioration may be increased in situations which stress the cardiovascular system. After opening a portocaval shunt, it is expected that a large amount of blood will be distributed from the splanchnic to the pulmonary circulation and put a strain on the heart. This two-dimensional echocardiographic study was made in 30 patients with cirrhosis and in 20 patients who were chronically treated (range: 16 to 156 months) with portasystemic shunt for prevention of hemorrhage from esophageal variceal bleeding. Patients with portasystemic shunts revealed a change in hemodynamic pattern. There was a significant increase in the left ventricular end-diastolic volume index and also a slight increase in the left ventricular end-systolic index. Cardiac output remained high despite a significant decrease in heart rate due to an elevated left ventricular stroke volume index. The parameters of systolic ventricular performance were normal. In contrast to the acute opening of the portacaval shunt, the chronic shunt volume put no strain of clinical significance on the heart.

Erythropoietin-dependent erythrocytosis associated with hepatic angiosarcoma

Erythropoietin (Ep) dependent erythrocytosis was discovered in a 77-year old man with angiosarcoma involving the liver and spleen. At autopsy, tissue from tumorous and normal sections of liver was obtained and extracts prepared for Ep assay in exhypoxic polycythemic mice. Neither tumor nor normal extracts had Ep activity; however, when the extracts were incubated with normal plasma prior to assay, significant amounts of Ep were generated. The normal liver tissue had three times the activity of the tumorous portion, 0.21 U/ml. These results suggest that the tissue contained an Ep activator, probably erythrogenin. It is proposed that the greater amount of erythrogenin activity in the normal tissue may reflect an attempt at liver regeneration following damage by an invading tumor, as regenerating liver has been identified as a stimulus for extrarenal EP and erythrogenin production.

Pulmonary hypertension in patients with pulmonary arteriovenous fistulae

Images