A new approach to facilitate diagnosis of nonalcoholic fatty liver disease through a galactose single point method in rats with fatty liver.

Background: Liver biopsy reliably diagnoses nonalcoholic fatty liver disease, but its invasiveness and inter- and intra-observer errors limit its usefulness in monitoring.

Aims: Use a galactose single point method or combined biochemical parameters to improve assessments of nonalcoholic fatty liver disease in a rat model.

Methods: Three nonalcoholic fatty liver disease severities were generated in 50 rats: a control group (n=18) on a standard diet, and 2 study groups on a choline-deficient diet (n=18), with and without treatment with silymarin (n=14). At weeks 4, 8, and 18, a galactose solution (0.5 g/kg/body weight) was rapidly injected intravenously. Sixty minutes later, internal artery blood was taken for biochemical analyses, including galactose. The livers were then removed for haematoxylin-eosin staining and to measure the hepatic lipid content.

Results: Alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, albumin, and total protein were each significantly correlated with nonalcoholic fatty liver disease severity. Regarding logistic regression, galactose single point method and total protein were significantly predictive. The optimal alanine aminotransferase cutoff point for nonalcoholic fatty liver disease prediction from the receiver-operating characteristic curve had 72.4% sensitivity and 52.4% specificity; galactose single point method alone had 82.8% and 72.4%, whereas galactose single point method+total protein showed 82.8% and 81.0%.

Conclusions: Both galactose single point method and galactose single point method+total protein had greater diagnostic sensitivity and specificity for nonalcoholic fatty liver disease than traditional biochemical tests.