AIMS:This study aims to evaluate a preclinical model of hepatic failure to propose an effective scheme to follow progression of mesenchymal cell treatments using serum parameters. METHODS: Male Wistar rats (n = 53) received dimethylnitrosamine (DMN) or PBS (control) intraperitoneally. We evaluated liver macroscopy, histology, serum albumin (ALB), total bilirubin (TB), alanine (ALT) and aspartate (AST) aminotransferase, alkaline phosphatase (ALP), gamma globulin, and hepatic gene expression of ALB and selected regeneration markers. RESULTS: After defining the effective DMN dose, 1 × 10 7 umbilical cord mesenchymal stem cells (UCMSC) were injected into the tail vein 1 week after treatment and liver function re assessed. DMN 10 µg/g delivered 3 days/week for 4 weeks altered liver macroscopy and histology. DMN treatment also increased ALP, ALT and TB, and significantly reduced gene expression of regeneration factors, although ALB and hepatic growth factor mRNA were not altered. Upon UCMSC treatment, a greater percentage of rats showed reduction of AST levels (75% UCMSC vs 0% PBS). CONCLUSIONS: Selected serum parameters can predict liver function in a rodent model, but each animal must be its own control.