AIM To understand the relationship between levels of endotoxin and cytokines in serum and to clarify thecause of cytokines change in liver diseases.METHODS Serum endotoxin level was determined by quantitative limulus amebocyte lysate chromogenicassay in 89 cases of acute and chronic liver diseases. Cytokines (TNF-a, IL-2, IL-6, IL-8, G-CSF) wereassayed by ELISA. Patients were divided into two groups based on oral administration of lactulose or not.Mean concentration of endotoxin and cytokines was compared before and 20 days after lactulose treatment.RESULTS The highest serum level of endotoxin was found in patients with cirrhosis (69.3±23.6pg/mL)and the lowest in patients with chronic hepatitis (28.4±7.9pg/mL), the moderate in patients with acutehepatitis (44.6±14.3pg/mL) (P<0.01). Serum levels of TNF-a, IL-2, IL-6, IL-8, G-CSF were higher inpatients with acute hepatitis than those with chronic hepatitis (P<0.05). No difference was noted betweenchronic hepatitis and cirrhosis (P >0.05). In all cases, serum levels of endotoxin were positively correlatedwith the concentration of TNF-a (r=0.555, P<0.05), IL-6 (r=0.531, P<0.01), IL-8 (r=0.440,P<0.05) and G-CSF (r =0.440, P<0.05), but not with IL-2 (r =0.10l, P<0.05). The decrease of serumlevels of endotoxin was greater in patients taking lactulose than controls (25.6±14.4pg/mL, n = 49 cases vs.10.9±9.Spg/mL, n = 40 cases, P < 0.01), the recovery from endotoxemia was higher in group withlactulose treatment than in controls (94.7%, n = 19 vs 36.4%, n = 22, P < 0.01 ). The decrease ofendotoxin resulted in decreases of TNF-a, IL-6, IL-8, G-CSF, ALT, AST and TB.CONCLUSION Endotoxemia is common in liver diseases, which could induce production and release ofcytokine from monocytes and macrophages and has harmful effects on hepatocytes. Treatment with lactulosecould decrease serum levels of endotoxin and cytokines, suggesting that lactulose could protect liver cells frominjury by reducing the absorption of endotoxin in intestine.

Objective: To test the hypothesis that acute phase reactants, such as alpha 1-antitrypsin and alpha 1-acid glycoprotein, could protect mammalian cells from further damage.Methods: Human dermal fibroblasts (5 × 104) were cultured with DMEM plus 10％ FBS at 37℃ in a 5％ CO2incubator. Different doses of LPS (lipopolysaccharide) and/or acute phase reactants were added. After 24 hours, the cultured supernatant was aspirated, the cells were washed, fixed and stained by methylene blue. The unbound stain was washed off. The stained cells were solubilized in 0.1 mi of 1％ Triton X-100. The absorbance of each well was measured using an ELISA spectrophotometer. The concentration of LPS which decreased the absorbance to 70％ of the control LPS-free ) cultures was defined as LD30.Results: In order to achieve LD30 in the presence of acute phase proteins, it was necessary to alter the LPS concentrations. The LD30 of LPS treated with 0, 0.5, 2, 10mg/ml antitrypsin and 0, 0.5, 2, 10 mg/ml glycoprotein was 5.4, 6.5, 7.6, 14.2 mg/ml and 5.2, 5.9, 6.9, 10.5mg/ml, respectively. Statistically, with the treatment of more than 2 mg/ml antitrypsin or glycoprotein, LD30increased significantly.Conclusions: Our data show that fibroblasts are susceptible to the direct toxicity of LPS. Alpha 1-antitrypsin and alpha 1-acid glycoprotein can reduce the toxicity and/or increase the tolerance of mammalian cells to LPS.