AIM: To determine whether serum hyaluronic acid reliably predicts the severity of hepatic fibrosis in patients with non-alcoholic fatty liver disease (NAFLD). METHODS: We studied 79 patients with histologically confirmed NAFLD. Hyaluronic acid was measured in serum obtained at the time of liver biopsy. Severity of fibrosis was staged based on Brunt's classification. The prediction levels for fibrosis were determined by area under the receiver operating characteristic curve (AUC). RESULTS: The logarithm of serum hyaluronic acid was significantly different among the stages of fibrosis (P<0.0001, analysis of variance) and had a significant positive correlation with the degrees of fibrosis after adjusting for age and serum albumin (partial r=0.44, P<0.0001). AUCs were 0.67, 0.87, 0.89, and 0.92 for any levels of fibrosis, > or =moderate fibrosis, > or =severe fibrosis, and cirrhosis, all of which were significantly higher than 0.5 (P<0.05). The cut-off value of serum hyaluronic acid of 46.1 mug/l was associated with the highest AUC for severe fibrosis, yielding a sensitivity of 85% and a specificity of 80%. The corresponding positive and negative predictive values were 51% and 96%, when assuming prevalence of severe fibrosis in NAFLD patients of 20% at referral centers. CONCLUSIONS: Measurement of serum hyaluronic acid is useful to identify NAFLD patients with severe fibrosis.
AIM: To determine whether serum hyaluronic acid reliably predicts the severity of hepatic fibrosis in patients with non-alcoholic fatty liver disease (NAFLD). METHODS: We studied 79 patients with histologically confirmed NAFLD. Hyaluronic acid was measured in serum obtained at the time of liver biopsy. Severity of fibrosis was staged based on Brunt's classification. The prediction levels for fibrosis were determined by area under the receiver operating characteristic curve (AUC). RESULTS: The logarithm of serum hyaluronic acid was significantly different among the stages of fibrosis (P<0.0001, analysis of variance) and had a significant positive correlation with the degrees of fibrosis after adjusting for age and serum albumin (partial r=0.44, P<0.0001). AUCs were 0.67, 0.87, 0.89, and 0.92 for any levels of fibrosis, > or =moderate fibrosis, > or =severe fibrosis, and cirrhosis, all of which were significantly higher than 0.5 (P<0.05). The cut-off value of serum hyaluronic acid of 46.1 mug/l was associated with the highest AUC for severe fibrosis, yielding a sensitivity of 85% and a specificity of 80%. The corresponding positive and negative predictive values were 51% and 96%, when assuming prevalence of severe fibrosis in NAFLDpatients of 20% at referral centers. CONCLUSIONS: Measurement of serum hyaluronic acid is useful to identify NAFLDpatients with severe fibrosis.