LAUSR

We would like to thank Tamaki et al. for their interest in and valuable comments on our study. As Tamaki et al. pointed out, we agree that nonalcoholic fatty liver disease (NAFLD) includes a non‐negligible number of patients at high risk for atherosclerotic cardiovascular disease (ASCVD). However, NAFLD is heterogeneous in terms of the concomitant known ASCVD risks such as diabetes, dyslipidemia, and hypertension. Thus, NAFLD can be classified into the “non‐metabolic” or “metabolic NAFLD” group according to the concomitant metabolic dysfunctions. The definition of metabolic dysfunction‐associated fatty liver disease (MAFLD) excludes the non‐metabolic NAFLD group, but not the entire NAFLD group. Accordingly, we should discuss if exclusion of the non‐metabolic NAFLD group has a negative impact on ASCVD‐related mortality. We enrolled subjects with fatty liver in our study and could not evaluate the difference in the ASCVD risk between the non‐ metabolic NAFLD and no fatty liver groups. However, recent studies examined the difference using the Third National Health and Nutrition Examination Survey (NHANES III). Kim et al. and Huang et al. demonstrated no significant difference in ASCVD‐ related mortality between the non‐metabolic NAFLD and no fatty liver groups, even using multivariate analysis adjusted for age, sex, race/ethnicity, and non‐invasive fibrosis indices (Table 1). In addition, no significant difference was observed in all‐cause mortality between the non‐metabolic NAFLD and no fatty liver groups in the multivariate analyses (Table 1). Therefore, the non‐metabolic NAFLD group seems to be comparable to subjects with no fatty liver in terms of ASCVD‐related mortality as well as all‐cause mortality. Tamaki et al. referred to our previous study, which found that NAFLD was also independently associated with significant fibrosis. However, the liver stiffness evaluated by shear ware elastography was within the normal range in all subjects in the non‐metabolic NAFLD group in our previous study (median 5.2 kPa). Recent studies also examined the prevalence of advanced fibrosis in the non‐metabolic NAFLD group. A analysis of four studies and our previous study demonstrates that hepatic fibrosis had an incidence of 0.8% in the non‐metabolic NAFLD group (5/594). Thus, the non‐metabolic NAFLD group seems to be at low risk; however, caution should also be used, as individuals with non‐metabolic NAFLD may develop MAFLD over time because metabolic risks increase with age. There is no doubt that liver fibrosis assessment is useful for identifying high‐risk patients even in the non‐metabolic NAFLD group, as Tamaki et al. suggested. However, fatty liver affects approximately 25% of the adult population and has emerged as a worldwide economic and health burden. In addition, accurate non‐invasive assessments for hepatic fibrosis such as vibration‐ controlled transient elastography and magnetic resonance elastography are not yet widespread. Moreover, no effective treatment has been established for NAFLD or hepatic fibrosis. On the other hand, metabolic dysfunction is a well‐established potent risk factor for intra‐ and extra‐hepatic events. MAFLD may be useful for identifying high‐risk patients at the large population level with no specific modalities and metabolic dysfunction is treatable by medication. Although MAFLD is still under debate and further studies are mandatory, a new definition of MAFLD may be a step toward renovating clinical practice for fatty liver.