Objective: The aim of this study was to evaluate cytokeratin-18M65 (CK-18M65) for distinguishing between simple steatosis (SS) and non-alcoholic steatohepatitis (NASH) against healthy individuals (HIs) in Japanese population. Methods: The serum from 24 HIs, 21 patients with SS and 20 patients with NASH were examined. Serum CK-18M65 was measured by enzyme-linked immunosorbent assay. Results: Aspartate aminotransferase was significantly different between NASH patients and HIs with p < 0.0001 (SS patients and HIs: p < 0.0001), as was alanine aminotransferase between NASH patients and HIs with p < 0.0001 (SS patients and HIs: p < 0.0001). Serum CK-18M65 increased in a stepwise fashion in HIs and also in SS and NASH patients. Multivariate logistic regression analysis revealed that NASH could be diagnosed with the use of CK-18M65 alone (p = 0.0285, OR 1.0038, 95% CI 1.0004-1.0073). At the optimal cut-off level of 548 U/l, CK-18M65 had an AUC value of 0.7369, 60.00% sensitivity and 85.70% specificity. In patients with NASH, no significant difference was observed between low fibrosis (Stage 0-1, 794.30 ± 454.41, n = 10) and high fibrosis (Stage 2-3, 809.70 ± 641.43, n = 10; p = 0.5967) and between slight steatosis (<33%, 512.89 ± 229.65, n = 9) and moderate steatosis (≥33%, 655.13 ± 480.78, n = 32) in patients with non-alcoholic fatty liver disease (NAFLD; p = 0.7647) with the use of CK-18M65. Conclusion: Serum CK-18M65 distinguished NASH from SS, but could not assess the severity of steatosis in NAFLD patients or the grade of fibrosis in NASH patients in Japanese population.