The occurrence of nonalcoholic fatty liver disease is a well-known side-effect of steroid therapy [1, 2]. In particular, treatment with high doses of glucocorticoid has been related to liver steatosis and more recently to nonalcoholic steatohepatitis (NASH) [3, 4]. We report a case of steatohepatitis induced by methylprednisolone therapy in a patient with an exacerbation of ulcerative colitis (UC).
A 25-year-old man with a 10-year history of UC was admitted for the occurrence of bloody diarrhoea (8–10 bowel movements a day) and abdominal pain. Examination showed abdominal tenderness. Three stool cultures for parasites and bacteria were negative. Blood examinations showed normal ALT (32 U L−1; normal values 5–45) and AST (27 U L−1; normal values 10–45), and increased ERS (45 mm, normal values <10 mm) and C reactive protein (CRP-22 ng mL−1; normal values <5 ng mL−1). An abdominal ultrasound examination was then performed with no evidence of liver diseases. Colonoscopy showed left colitis with marked mucosal hyperaemia with small punctate ulcers, and biopsies showed severe chronic inflammation with presence of multiple crypt abscesses, epithelial flattening and ulceration consistent with UC exacerbation. Methylprednisolone therapy 40 mg day−1 showed an improvement in symptoms within a few days with two daily bowel movements of formed stool without blood. Then he was dismissed. Therapy was tapered to 16 mg in 4 weeks. After 40 days blood examination showed an increase in ALT value (106 U L−1). AST remained normal (41 U L−1). ERS and CRP were normal (8 mm at first hour and 3 ng mL−1, respectively). An abdominal ultrasound examination was then repeated with the evidence of fat infiltration consistent with liver steatosis.
Hepatitis virus (HAV, HBV, HCV, HEV, HGV, CMV and EBV) and antibodies against mitochondria, smooth muscle and nuclear antigens were not detected by serological tests. Liver biopsy showed macrovescicular hepatocellular fat accumulation, periportal inflammation and mild fibrosis (Fig. 1). The patient denied a history of alcohol abuse and the intake of hepatotoxic drugs and a diagnosis of nonalcoholic steatohepatitis was made . He stopped the steroid treatment in 2 weeks and ALT value gradually decrease to normality. An abdominal ultrasound examination performed after 3 months showed an improvement in liver morphology with reduction in fat infiltration. Serological test for virus and autoantibodies were retested in the hypothesis of previous false negative results due to steroid therapy and were confirmed negative. The patient refused to participate in a second liver biopsy. After 6 months abdominal ultrasound showed a normal liver. Now, he is in remission with a 5-ASA maintenance therapy, with normal values of ALT and AST.
Previously, steroid treatment has been related to steatohepatitis and fatal liver failure in two patients affected by systemic lupus erythematosus  and in our case steatohepatitis appears after starting steroid treatment in UC. Although drug discontinuation seems effective to induce remission of this condition a careful follow-up of patients submitted to steroids has to be performed because of the possibility of development of liver failure.