In their review of NASH, Sheth and colleagues [1] reaffirmed the proposed diagnostic criteria for this condition; liver biopsy specimens showing moderate to gross macrovesicular fatty infiltration with inflammatory processes in the absence of serologic evidence of known viral hepatitides and substantial alcohol consumption.

Conditions almost always associated with NASH include obesity, diabetes mellitus, and hyperlipidemia (hypercholesterolemia, hypertriglyceridemia, or both). However, the temporal relation between NASH and diabetes mellitus has not been described previously. While evaluating patients with chronically elevated liver function test results, we noted that the diagnosis of NASH preceded the diagnosis of diabetes mellitus in some patients. The mean age of the four patients was 45 years, the mean triglyceride level was 285 mg/dL, and the mean cholesterol level was 265 mg/dL. The mean lag between diagnosis of NASH and subsequent diabetes mellitus was 10 months (range, 5 to 18 months). Eighty percent of patients were male, and all patients were white, occasionally used alcohol, and were 20% to 30% above their ideal body weights. None had previous evidence of hyperglycemia at the initial diagnosis of NASH. All patients had normal thyroid function test results, diagnostic liver biopsy specimens consistent with NASH, and no other known causes of hepatitis. All patients had abnormal computed tomographic scans, which showed fatty infiltration.

It is interesting that hyperlipidemia, obesity, and elevated insulin levels are some of the features shared by NASH and diabetes mellitus [2]. Insulin inhibits oxidation of free fatty acids, enhancing the toxic effects of these acids in the liver [3]. We therefore recommend that patients with NASH and euglycemia be followed closely for the development of type 2 diabetes mellitus.