Abstract :
Background: Non-alcoholic fatty liver disease (NAFLD) is a disorder which ranges over a spectrum, extending from simple steatosis to steatohepatitis (NASH). Significant proportion of NASH could progress to cirrhosis. However, it has remained difficult to differentiate between NASH and simple steatosis by clinical examination. Nowadays serum high sensitivity C-reactive protein (hs-CRP) is an inflammatory marker that predicts severity of metabolic syndrome.
Aim: To study whether serum hs-CRP including other metabolic parameters could differentiate and predict NASH in patients with NAFLD.
Methods: Seventy-six patients with biopsy proven NAFLD (mean age 53 ± 10.5), 15 with simple steatosis (mean age 49.5 ± 11.3) and 61 with NASH (mean age 53.9 ± 10.2) were investigated. Baseline characteristics including BMI and waist-hip ratio of all subjects were evaluated. Serum fasting plasma glucose, lipid, hs-CRP, leptin, adiponectin, TNF-α and Homeostasis Model Assessment Method (HOMA-IR) were measured. Metabolic syndrome was assessed according to National Cholesterol Program’s Adult Treatment Panel III. Liver biopsies were graded according to Brunt criteria. Diagnostic accuracy of predictors of NASH and advance fibrosis was examined prospectively.
Results: The median (min-max) of serum hs-CRP was higher in NASH compared with simple steatosis [1.9 (0.2-14.2) vs 0.97 (0.2-12.3) mg/L]. And the optimal cut-off value of serum hs-CRP in differentiating NASH and simple steatosis, assessed by ROC analysis was 0.88 mg/L, with sensitivity of 85.3%, specificity of 46.7%, and accuracy of 88.1%. In univariate analysis, factors associated with NASH were female sex, presence of metabolic syndrome criteria > 2, FPG, HOMA-IR and serum hs-CRP ≥ 0.88 mg/L. In multivariate analysis, serum hs-CRP ≥ 0.88 mg/L (OR 4.9; 95% CI 1.3-18.5 p = 0.02) and presence of metabolic syndrome criteria > 2 (OR 5.0; 95% CI 1.3-19.8 p = 0.02) were independent predictors of NASH. When both factors were concomitantly positive, the accuracy to detect NASH as determined by ROC analysis was 0.7 (95% CI 0.6-0.9) and had an accuracy of 82.9%.
Conclusions: Serum hs-CRP is a useful non-invasive marker that differentiates between NASH and simple steatosis in NAFLD patients. |
file/Thai-Journal-of-gastroenterology-vol-11-no-3-1323350.pdf