Insulin resistance in nonalcoholic steatohepatitis: necessary but not sufficient - death of a dogma from analysis of therapeutic studies?

Studies on pathogenesis tend to blame insulin resistance as the chief pathogenic agent in the development and progression of nonalcoholic steatohepatitis (NASH). In this article, studies reporting histological changes induced by pharmacological therapy and nonpharmacological interventions in NASH are critically reviewed, assuming that analysis of morphological findings can provide further insight into the pathogenesis of NASH. PubMed database analysis provided 16 studies describing light microscopy in adults and three in children; ultrastructural analysis was conducted through electron microscopy in two human and four animal studies. Analysis of the data disclosed methodological issues, such as variable histological criteria, limited series, failure to stratify enrolled patients for their risk of progression and very few electron microscopy studies. Moreover, no particularly convincing 'proof-of-concept' study that might assist in understanding the pathogenesis of NASH was found. It is noteworthy that insulin sensitizers fail to treat NASH in all cases, do not reverse or even worsen mitochondrial abnormalities in NASH and, conversely, histological improvement of disease, at least in some patients, is observed with agents acting through mechanisms other than insulin sensitization, such as vitamin E. The finding that correction of insulin resistance may not be sufficient to successfully treat NASH in the majority of patients seems to conflict with studies on pathogenesis. This might imply that NASH is the shared end result of varying pathogenic mechanisms concurring to determine liver damage to a variable extent in the individual patients. If this hypothesis is true, we should try to tailor treatment to each subject.