Association of Serum Homocysteine Levels with Histological Severity of NAFLD

Background & Aims: Studies on the association between homocysteine and non-alcoholic fatty liver disease (NAFLD) have shown inconsistent results. Our study concerns the association of homocysteine with the histological severity of NAFLD, especially non-alcoholic steatohepatitis (NASH) and significant fibrosis (SF) after adjusting for other well-identified risk factors. Methods: This study enrolled 289 patients with biopsy proven NAFLD. The association of homocysteine with the severe histological features was examined using multivariable logistic regression analysis and subgroup analysis. The area under curves (AUC) and Hosmer-Lemeshow goodness-of-fit test for the adjusted logistic regression models was analyzed. Results: After multivariable regression analysis, homocysteine showed significant correlation with NASH (OR 0.79 95%CI: 0.69-0.89), p<0.001) and SF (OR 0.83 95%CI: 0.72-0.95, p=0.009). Spearman's correlation analysis showed homocysteine levels were inversely correlated with the grade of hepatocellular ballooning and the stage of liver fibrosis (Spearman's.=-0.13, p=0.033; Spearman's.=-0.16, p=0.007), but had no correlation with the severity of steatosis and lobular inflammation. The subgroup analyses showed that homocysteine was strongly associated with NASH in females but was weaker in males (female OR: 0.61 95%CI: 0.45-0.84; male 0.86 95%CI: 0.75-0.99), and on SF showed no significant differences in the subgroups. The models showed good discrimination for NASH (AUC 0.789, 95% CI: 0.736-0.843) and for SF (0.784 95%CI: 0.719-0.848) and calibration (Hosmer-Lemeshow goodness-of-fit test, p=0.346 for NASH; p=0.908 for SF). Conclusion: Elevated serum homocysteine levels are negatively associated with NASH and SF in subjects with NAFLD.