LAUSR

We read with much interest the recently published study by Thuluvath et al. questioning the hypothesis that patients with cryptogenic cirrhosis listed for liver transplantation may actually represent the evolution of patients with nonalcoholic fatty liver disease (NAFLD), as well as the accompanying editorial article by Drs. Caldwell and Marchesini who critically reassessed the similarities and differences between non-alcoholic steatohepatitis (NASH)-related and cryptogenic cirrhosis. We do agree with the vision expressed in the editorial article, as we feel that the majority of patients who, in the past, were labelled as having ‘‘cryptogenic cirrhosis” were actually affected by burned-out NASH, wherein some of the main clinical characteristics of NAFLD, such as diabetes mellitus and increased body mass index, were disguised by increased prevalence and by modification in body composition due to fluid retention in advanced liver disease, respectively. Concurrently, other characteristics such as central adiposity, were diluted by the loss of both lean and fat mass resulting from the complex metabolic alterations that often occur in patients with end-stage liver disease. As a fact, although the authors of the article surmise that cryptogenic cirrhosis should not be equated with the term ‘NASH cirrhosis’, because of the presence of differences between these two entities even when patients were matched for some demographic variables (i.e., age, gender), year of listing for liver transplantation, or when patients were stratified by some of the typical clinical features of NASH such as presence of diabetes mellitus and obesity, we feel that the retrospective evaluation of the United Network for Organ Sharing (UNOS) database does not do justice to the similarities that we feel exist between cryptogenic cirrhosis and NASH. In particular, although the authors pointed out that the presence of diabetes mellitus, in the UNOS database, is significantly over-represented in patients with NASH cirrhosis compared to the other aetiologies evaluated in the study (cryptogenic cirrhosis, alcoholic cirrhosis, autoimmune hepatitis), if considering all the periods of evaluation – 2002–2016 and 2011–2016 – there is clear evidence from the data provided in the manuscript that the prevalence of this metabolic condition is significantly higher in patients with cryptogenic cirrhosis (2002–2016, 28.5%; 2011–2016, 31.9%) than in those with alcoholic cirrhosis (2002–2016, 16.1%, p < 0.0001; 2011–2016, 18.0%, p < 0.0001) and autoimmune hepatitis (2002–2016, 16.3%, p < 0.0001;