We congratulate Fracanzani and colleagues 1 on their insightful article identifying factors associated with severe liver disease in patients with nonalcoholic fatty liver disease (NAFLD) and normal aminotransferase levels. First, the authors have convincingly demonstrated that normal alanine aminotransferase (ALT) is not a valuable criterion to exclude patients with NAFLD from liver biopsy. More importantly, they have shown that increased insulin resistance as assessed by the homeostasis model assessment of insulin resistance (HOMA-IR) is a statistically significant and independent predictor of the presence of both nonalcoholic steatohepatitis (NASH) and severe fibrosis among subjects with NAFLD. The authors concluded that determination of insulin resistance in subjects with normal ALT levels may be clinically relevant in the selection of NAFLD cases for histological assessment of disease severity and progression. 1 The field of biomarkers is an area of fast growing interest in the setting of NAFLD. A neoepitope in cytokeratin 18, termed M30 antigen, becomes available at an early caspase cleavage event during apoptosis and has been regarded as a biochemical marker of liver injury. 2,3 Thus, we sought to determine whether serum M30 levels may be associated with the presence of NASH among NAFLD cases with normal ALT levels. Eighteen patients (four males and 14 females), mean age of 51.0 Ϯ 7.6 years, with liver biopsy-confirmed NAFLD and normal ALT levels (Ͻ 40 U/L) were investigated. The diagnosis of NASH was based on the Kleiner criteria. 4 M30 levels were detected by enzyme-linked immunosorbent assay (M30 Apoptosense ELISA kit; Peviva AB, Bromma, Sweden). Compared with patients with NAFLD without NASH (n ϭ 12), mean serum M30 levels were significantly raised in the six patients with normal ALT and NASH levels (198.2 Ϯ 37.2 IU/L versus 80.7 Ϯ 19.2 IU/L, P Ͻ 0.01). In multivariate analysis, M30 levels and HOMA-IR were the only independent predictors of the presence of NASH in patients with NAFLD and normal ALT levels.Besides confirming the elegant proof by Fracanzani and coworkers that determining HOMA-IR is clinically useful for identifying patients with NAFLD who are candidates for histological assessment of disease severity, 1 our pilot results allow us to postulate that M30 levels may be an additional good index of the presence of NASH in this patient group. Further studies in larger samples are warranted to confirm our pilot data.
Reply:We thank Dr. Yilmaz et al. for comments on our article. 1 The results of the pilot study Yilmaz and colleagues conducted in patients with nonalcoholic fatty liver disease (NAFLD) and normal alanine aminotransferase (ALT) levels confirm our finding that the severity of insulin resistance is an independent predictor of liver damage in these patients, and suggest that serum levels of M30, a neoepitope of cytokeratin 18 (CK-18) released during the early phases of apoptosis, may prove useful to discriminate patients with nonalcoholic steatohepatitis (NASH) from those with simpl...