• diabetes
• fibrosis
• hepatitis C
• non-alcoholic steatohepatitis
• steatosis

We read with great interest the article by Castera and colleagues (Gut 2003;52:288–92) and acknowledge the finding that worsening of steatosis in chronic hepatitis C is associated with fibrosis progression. However, in our view there are no data supporting a casual role for this statistical association or any specific relation of this finding to chronic hepatitis C.

Firstly, the authors provide no explanation as to why steatosis worsened in patients under consideration. Overweight, diabetes, and alcohol consumption are the main causes of steatosis in Western countries and major causes of fibrotic liver disease. There are no data throughout the study indicating whether patients in whom steatosis worsened simply gained weight or developed any of the complications associated with insulin resistance. The latter can develop within the course of liver injury well before cirrhosis is present¹ or be epidemiologically linked to infection by hepatitis C virus (HCV) for reasons that have yet to be determined.² High serum glucose,³ as well as diabetes,⁴ are associated with liver fibrosis progression⁵ and might contribute to enhanced fibrogenesis.⁶ As for alcohol consumption, a thorough evaluation is needed before ruling out the possibility of even slight increases in daily alcohol consumption translating, over the course of several years, into enhanced steatosis. There is a theoretical possibility that progression of steatosis reflects the natural course of HCV infection if steatosis were to occur later than the necroinflammatory lesions defining chronic hepatitis. However, as current knowledge stands, this is purely speculative and also, there is no indication in this study that patients in whom steatosis progressed had a longer duration of infection than those in whom it did not. Hence there appears to be no data in this study suggesting that progression of steatosis is HCV-related or that confounding prosteatogenic factors have been ruled out.

The second issue is that it has not been made entirely clear what “worsening” of steatosis means. This was defined as an increase of at least one point on a grading scale that is not evenly distributed (0%; 1–10%; 10–30%; >30%). Since many patients had no steatosis on the first biopsy, such a definition would mean that in most cases an increase from 0% to 5% would be qualified as “worsening” of steatosis. This may explain the authors’ statement that “there were less patients with progression of steatosis than patients with steatosis appearance between the two biopsies”. In any event, the biological relevance of minor increases in the amount of steatosis appears highly improbable, especially if the total amount of steatosis on the first biopsy was not associated with the amount of fibrosis, as noted in this study. This biological relevance could have been strengthened had the authors provided quantitative data on a correlation between progression of steatosis and progression of fibrosis.

Although the idea that steatosis progression rather than the amount of steatosis is associated with fibrosis progression warrants further study, it is hard to reconcile with lessons from non-alcoholic fatty liver disease where patients with massive steatosis do not develop liver fibrosis⁷ although they obviously experienced steatosis progression. This argues against a simple and direct link between steatosis and fibrosis. We propose an alternate view in which both steatosis and fibrosis are the result of a common underlying condition, insulin resistance, which operates through proinflammatory mediators⁸ to enhance fibrogenesis and through alterations in metabolic pathways to promote steatosis.