Abstract
AIMS/INTRODUCTION: To explore the association between estimated small dense low-density lipoprotein cholesterol (sdLDL-C) and the risk of incident nonalcoholic fatty liver disease (NAFLD) in nonobese populations. MATERIALS AND METHODS: This study included participants who underwent health checkups in 2014 and were followed up until 2019. We carried out Cox proportional hazards regression analyses to evaluate the association of estimated sdLDL-C with NAFLD. Discordance analyses were carried out to estimate the relative NAFLD risk in estimated sdLDL-C versus low-density lipoprotein cholesterol (LDL-C) discordant/concordant groups. Estimated sdLDL-C was calculated by equations based on LDL-C and triglycerides. The diagnosis of NAFLD was based on the presence of abdominal ultrasonography after excluding other causes of chronic liver disease. RESULTS: Over a mean follow-up period of 26,694 person-years, 844 incident NAFLD cases were recorded. Compared with the first quartile of estimated sdLDL-C, the fourth quartile was associated with a 2.933-fold increased risk of NAFLD (95% confidence interval 2.095-4.107). With the increase in estimated sdLDL-C, the risk of NAFLD gradually increased both in participants within the normal range of LDL-C (hazard ratio 2.854, 95% confidence interval 1.650-5.617) and beyond the normal range of LDL-C (hazard ratio 2.636, 95% confidence interval 1.263-5.502). In addition, the inconsistent high estimated sdLDL-C/low LDL-C group was associated with an increased risk of NAFLD, but not the low estimated sdLDL-C/high LDL-C group. CONCLUSIONS: Estimated sdLDL-C was positively associated with the risk of incident NAFLD in a nonobese population, independent of LDL-C.