Aims: Traditionally, sarcopenia has defined as amount of absolute muscle mass adjusted by height in the elderly people. However, relative muscle mass adjusted by weight has used extensively in most non-alcoholic fatty liver disease (NAFLD) studies. Here, we attempted to investigate the pitfall of sarcopenia in NAFLD according to adjustment methods. Methods: Adult subjects (n=1,343) who underwent a health check-up at the healthcare center were finally included for analysis, except for 682 subjects. Total skeletal muscle (TSM), appendicular skeletal muscle (ASM) and fat measured using the bioelectrical Impedance Analysis (BIA). The weight-adjusted skeletal muscle mass index (wSMI) and height-adjusted SMI (hSMI) calculated by dividing the total ASM by weight or the square of height, respectively. Fatty liver diagnosed by using abdominal sonography. Results: Prevalence of sarcopenia defined by wSMI in the NAFLD group was significantly higher than in the control group (1.3% vs. 8.8%, P<0.001). But there was no difference in the prevalence of sarcopenia defined by hSMI between the control and NAFLD groups (2.0% vs. 0.8%, P=0.055). The concordance rate of the two methods was 0.29% (4/1343) NAFLD. Because body weight was most potent independent risk factor for NAFLD in multivariable logistic regression analysis, when we adjusted parameters adjusted by body weight, prevalence of abnormality of almost all parameters increased in NAFLD population. When the bilirubin (or platelet) adjusted by weight, prevalence of abnormal bilirubin (or platelet) increased in NAFLD subjects. However, prevalence of abnormality of non-metabolic parameter (bilirubin or platelet) did not increase in NAFLD, after adjusting by height. Only metabolic parameters showed relationship with NAFLD, after adjusting by height Conclusions: The concordance rate of sarcopenia defined by wSMI and hSMI was very low. As NAFLD is highly associated with body weight, attention should be given in the case of studying the relationship of NAFLD with sarcopenia adjusted by body weight.