The thickening of the gallbladder wall in patients with ascites is commonly related to hypoalbuminemia and/or portal hypertension. To evaluate the pathogenetic role of these two factors, we correlated gallbladder wall thickness (GBWT) with the albuminemia and the serum-ascites albumin gradient (SAAG), an index of portal hypertension, in 47 patients with ascites caused by cirrhosis of the liver or abdominal malignancy. We found a thickened gallbladder wall in 30/47 patients. The correlation between GBWT and SAAG was 0.64 (n = 47 p less than 0.001). No correlation was found between GBWT and albuminemia (r = 0.04). We suggest that the sonographic finding of ascites and gallbladder wall thickening should be considered a valuable sign of transudative ascites and of portal hypertension whatever its cause.

Ticlopidine, a new antithrombotic agent, and theophylline, a widely used bronchodilator drug, are both almost completely metabolized in the liver. To evaluate an interaction between these two drugs, we studied theophylline pharmacokinetics before, after 10 days of ticlopidine administration, and 1 month later in 10 healthy volunteers. We found a highly significant increase in the theophylline elimination half-life (P less than 0.001) and a comparable reduction in its total plasma clearance (P less than 0.001) after ticlopidine treatment. Pharmacokinetic parameters returned to initial values within 30 days after ticlopidine withdrawal. Moreover, no changes in theophylline pharmacokinetic parameters were observed 3 months later, before and after 10 days of placebo administration. Our results seem to exclude direct liver toxicity and may suggest a reversible inhibition of the liver metabolic capacity of theophylline.

The plasma clearance after oral administration of a completely absorbed drug that is metabolized by the liver depends on its intrinsic clearance. In cirrhosis the bioavailability of a flow-dependent drug increases because of both portosystemic shunting and hepatocyte dysfunction. A drug with a high extraction ratio, lidocaine, and a drug with a low extraction ratio, theophylline, were administered to 27 patients with cirrhosis and 16 control subjects. We found a significant impairment of both theophylline clearance (p less than 0.001) and lidocaine clearance (p less than 0.001) and an increase in the lidocaine peak concentration (p less than 0.001). The three parameters were significantly correlated with each other. The impairment of theophylline metabolism did not correlate with other indexes of disease severity, whereas lidocaine clearance was lower and lidocaine peak level higher in patients with decompensated cirrhosis and evidence of portal hypertension. Thus impairment in lidocaine disposition, which reflects both hepatocyte dysfunction and portosystemic shunting, correlated closer with the severity of liver disease than did theophylline metabolism.

To evaluate the diagnostic accuracy of fibronectin levels in ascites to differentiate malignant from non-malignant ascites, the authors studied 30 patients with sterile uncomplicated ascites in chronic liver disease, 18 patients with malignant ascites and four patients with spontaneous bacterial peritonitis. Fibronectin concentration was significantly higher in malignant ascites than in sterile ascites (P less than 0.001). High values (greater than 85 mg/l) were found in four of six cases of hepatocellular carcinoma in liver cirrhosis with negative cytologic examination, and in six of seven peritoneal carcinomatoses. Low values (less than 85 mg/l) were found in four patients with liver metastases and in one patient with intrahepatic biliary duct carcinoma in cirrhosis. In four patients with infected ascites, the fibronectin level was low. Among all other parameters (total protein concentration, lactate dehydrogenase, gamma-glutamyl-transpeptidase, pH, amylase, triglycerides, leukocyte count, and cytologic examination), fibronectin yielded the best degree of discrimination (diagnostic accuracy, 79%).

The authors evaluated the diagnostic accuracy of sialic acid and its lipid-bound fraction in ascites and compared these tests with others (fibronectin, cholesterol) recently claimed as valuable in the differential diagnosis of ascites. Fibronectin yielded the best diagnostic accuracy (85%) with no false-positive and 37% of false-negative (10/27). The authors also found higher concentration of sialic acid in malignant ascites than in nonmalignant ascites (P less than 0.001) and, taking 300 mg/l as the cutoff value, the false-positive rate was 10% (four of 40), the false-negative rate 30% (eight of 27), and the overall diagnostic accuracy 82%, comparable to that of the fibronectin. The authors conclude that both fibronectin and sialic acid determinations in ascites may be regarded as accurate markers of neoplastic involvement of the peritoneum, although no test is useful in the ascites with hepatocellular carcinoma and cirrhosis of the liver.