LAUSR

INTRODUCTION We aimed to determine whether lobular inflammation and ballooning grades in the Non-alcoholic Steatohepatitis Clinical Research Network scoring system (NASH CRN) can be directly translated into the same for the Steatosis Activity Fibrosis scoring system (SAF), and to look at intra- and inter-observer agreement for each individual histological component and for diagnosis of non-alcoholic steatohepatitis (NASH) using the two scoring systems. METHODS Four pathologists from two Asian centres scored 20 digitalized slides, twice using the NASH CRN, twice using the SAF. Intra- and inter-observer agreement was analyzed using Fleiss' kappa, weighted kappa or Cohen kappa, where appropriate. RESULTS The intra-observer discrepancy rate when using the NASH CRN compared with the SAF was higher than when using the individual scoring system for lobular inflammation (15% comparing both scoring systems vs. 10% and 1.8% for the NASH CRN and the SAF, respectively) and hepatocyte ballooning (33.8% vs. 12.5% and 5%, respectively), but not for diagnosis of NASH (6.3% vs. 6.3% and 0%, respectively). Intra- and inter-observer agreement was substantial to almost perfect, except for inter-observer agreement for lobular inflammation and diagnosis of NASH, which was only fair to moderate in most instances. CONCLUSION These findings do not support the direct inter-translation between the NASH CRN and the SAF. However, the diagnosis of NASH during examinations using the NASH CRN may be comparable with diagnosis of NASH using the SAF, vice versa. The inter-observer agreement for lobular inflammation and NASH diagnosis needs to be improved.