CD14 is a
pattern recognition receptor constitutively expressed in different types of immune cells, either in a membrane-anchored (mCD14) or in a soluble (
sCD14) form. This study investigated whether hepatic CD14 expression levels were correlated with the grades of liver
inflammation as well as the potential usefulness of serum
sCD14 as a
biomarker for predicting liver
inflammation in
chronic hepatitis B (CHB) patients with normal or mildly elevated ALT.
METHODS: A total of 216 treatment-naive CHB patients with normal or mildly elevated ALT who underwent liver biopsy were recruited. Hepatic expression level of CD14 was measured using immunohistochemistry and real-time PCR. Serum
sCD14 concentrations were determined with an
enzyme-linked
immunosorbent assay. Correlations between hepatic CD14, serum
sCD14, and liver
inflammation grade were analyzed. Univariate and multivariate analysis were performed to identify significant liver
inflammation-associated factors. The receiver operating characteristic curve was used to assess the discriminating power of serum
sCD14 to significant liver
inflammation in CHB patients with normal or mildly elevated ALT.
RESULTS: Both hepatic expression levels of CD14 and serum
sCD14 concentrations significantly increased with the aggravation of liver
inflammation. Moreover, hepatic expression levels of CD14, serum
sCD14 concentrations, and liver
inflammation grades were positively correlated with each other. Three parameters including
alkaline phosphatase (ALP), neutrophil, and
sCD14 were identified as independent predictors of significant liver
inflammation. Subsequently, a diagnostic equation named model-sCD14 was developed incorporating
sCD14 and other variables (ALP and neutrophils) with p < 0.05 in multivariate logistic analysis. The area under receiver operating characteristic curve (AUC) of
sCD14 for predicting significant liver
inflammation was 0.788 and the optimal cutoff was 27.14 ng/mL, with a sensitivity of 66.67%, a specificity of 81.70%, positive predictive value of 60.01%, and negative predictive value of 85.62%. When
sCD14 was replaced by model-sCD14, the AUC value increased from 0.788 to 0.843 (z = 2.311, p = 0.021), with sensitivity of 77.78%, specificity of 77.12%, positive predictive value of 58.33%, and negative predictive value of 89.39%.
CONCLUSIONS: