Increased intercellular adhesion molecule-1 serum concentration in cholestasis

Increase of serum levels of the soluble intercellular adhesion molecules in patients with the cholestatic liver diseases primary biliary cirrhosis (PBC) and primary sclerosing cholangitis (PSC) are known and have been thought to indicate activation of the immune system and the grade of the inflammatory process. In hepatitis and cholestatic diseases, expression of adhesion molecules was found on the surface of bile duct epithelia and hepatocytes.

Serum levels of sICAM-1 in patients with intrahepatic cholestasis in PBC (n = 42) and extrahepatic cholestasis (n = 18) due to choledocholithiasis were investigated. sICAM-1 levels and "classical" cholestasis parameters as alkaline phosphatase (ALP), gamma-glutamyl-transpeptidase (gamma-GTP) and bilirubin levels were compared. Furthermore, sICAM-1 concentrations and "classical" cholestasis parameters were analysed before and after therapy with ursodeoxycholic acid (UDCA). In addition, sICAM-1 was detected in serum and bile fluid of four patients with cholestasis due to choledocholithiasis. Soluble ICAM-1 levels in sera and, if accessible, in bile fluids were determined using a commercially available ELISA system. Statistics were done by Wilcoxon's signed rank exact test and Spearman's rank correlation test. Sensitivity and specificity of cholestasis parameters and sICAM-1 concentrations was analysed by receiver operating characteristic (ROC) curves.

Increased sICAM-1 serum concentrations in a similar range were found in patients with PBC (range 251-2620 micrograms/l; median 966 micrograms/l) as well as in patients with extrahepatic cholestasis (257-2961 micrograms/l; median 760 micrograms/l) compared to healthy controls (n = 12; 220-500 micrograms/l; median 318 micrograms/l). sICAM-1 levels correlated significantly to histological stage I to IV (p < 0.001), ALP (range 107-1877 U/l; median 545 U/l; r = 0.496, p = 0.0008), bilirubin (range 0.3-26 mg/dl; median 0.8 mg/dl; r = 0.52; p < 0.0004) and gamma-GTP levels (range 43-705 U/l; median 221 U/l; r = 0.36; p = 0.02) in PBC patients. In PBC patients a histological stage III or IV (n = 21) could be predicted with high sensitivity (95%) and specificity (85%) if sICAM-1 levels were above 840 micrograms/l. After treatment of PBC patients with UDCA, sICAM-1 levels decreased significantly with decline of other "classical" cholestasis parameters. Increased sICAM-1 levels (range 257-2961, median 745 micrograms/l) in extrahepatic cholestasis correlated also significantly with serum concentrations of bilirubin (r = 0.8; p < 0.01; range 0.3-19.7, median 1.6 mg/dl), gamma-GTP (r = 0.55; p = 0.03; range 33-1401, median 179 U/l) and ALP (r = 0.61; p = 0.1; range 110-1378, median 562 U/l). sICAM-1 was detectable in bile fluid (264-919 micrograms/l) of four patients with extrahepatic cholestasis and nose-biliary catheterisation.

sICAM-1 concentrations were found to discriminate between histological stage I/II and stage III/IV of PBC with higher sensitivity and specificity than "classical" cholestasis parameters. Increased serum concentrations for sICAM-1 in intra- and in extrahepatic cholestasis and detection of sICAM-1 in the bile may indicate that sICAM-1 is eliminated through the bile. In other words, not only increased synthesis but also decreased elimination may be responsible for increased sICAM-1 serum levels in patients with cholestatic liver diseases.