Background and aimsThe relationship of non-alcoholic fatty liver disease (NAFLD) with the atherogenic index of plasma (AIP) is unclear. This study aims to detect the association between AIP and NAFLD, compare the discriminative power of AIP with other lipid parameters for NAFLD, and establish a discriminant model using physical examination data.MethodsParticipants aged over 20 years who underwent routine physical examination in Beijing Chaoyang Hospital from April 2016 to August 2020 were included. We categorized subjects based on hepatic ultrasound results and analyzed the association between NAFLD risk and AIP, conventional plasma lipids, remnant cholesterol (RC), triglyceride and glucose (TyG) index, and other atherogenic indices (n = 112,200) using logistic regression, restricted cubic spline regression, and receiver operating characteristic curve.ResultsOut of the 112,200 subjects, 30.4% had NAFLD. The body weight index, plasma glucose, conventional lipids, TyG index, AIP, atherogenic coefficient (AC), and coronary risk index (CRI) were significantly higher, while HDL-C was lower (p < 0.001) in patients with NAFLD than those without NAFLD (all p < 0.001). Compared with conventional lipids, RC, TyG index, AC, and CRI, AIP had a stronger correlation with the risk of NAFLD (OR 6.71, 95% CI 6.23–7.22, p < 0.001) after adjusting confounders and presented a non-linear dose–response relationship (p < 0.0001). The optimal cut-off value of AIP was 0.05 and the area under the curve (AUC) was 0.82 (95% CI: 0.81–0.82) with high sensitivity and specificity. The AUC of the simplified three-variable NAFLD discriminant model was 0.90 in both the training set and the validation set.ConclusionAIP was significantly associated with NAFLD and showed superior discriminative performance to other lipid parameters. These findings might help screen NAFLD in high-risk individuals and reduce the prevalence of NAFLD.

BackgroundHomocysteine (Hcy) is an independent risk factor for cardiovascular disease, while mechanisms are unclear. Despite inconsistent and limited, epidemiological and experimental studies indicated that hyperhomocysteinemia (HHcy) affected lipid metabolism. This study aims to investigate the association of plasma Hcy with traditional lipid profiles and remnant cholesterol (RC) in Chinese adults.MethodsIn total, 7,898 subjects aged 20–79 years who underwent a physical examination at Beijing Chao-Yang Hospital in Beijing were included in this study. Fasting plasma total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), apolipoprotein A1 (ApoA1), apolipoprotein B (ApoB), lipoprotein (a) [Lp(a)], Hcy, and other metabolic risk factors were measured by routine automated laboratory methods. RC was calculated as TC minus HDL-C and LDL-C. The linear regression model and logistic regression model were used to assess the relationship between Hcy and lipids after adjusting potential confounders.ResultsOf the subjects, the median level of plasma Hcy was 13.0 μmol/L and 32.3% had HHcy. Plasma Hcy was negatively associated with HDL-C, ApoA1, and Lp(a) and positively associated with TG levels after adjusting age, sex, body mass index, blood pressure, alanine transaminase, aspartate transaminase, creatinine, uric acid, and glucose. HHcy significantly increased the risk of low HDL-C [odds ratio (OR) 1.26; 95%CI (1.11–1.44); p < 0.001]. The net mediation effects of ApoA1 on the relationship between Hcy and HDL-C before and after adjusting confounders were 46.9 and 30.6%, respectively. More interestingly, the RC level was significantly elevated in subjects with HHcy after adjusting other influencing factors (p = 0.025). Hcy presented a positive correlation with RC levels after adjusting the above confounding factors (β = 0.073, p = 0.004), and the correlation was still significant even after controlling other lipids, including TG, LDL-C, HDL-C, ApoA1, ApoB, and Lp(a).ConclusionOur study showed that plasma Hcy was not only significantly associated with conventional atherogenic lipids but also independently correlated with RC levels beyond other lipids after controlling potential confounders. This finding proposes that identifying Hcy-related dyslipidemia risk, both traditional lipids and RC residual risk, is clinically relevant as we usher in a new era of targeting Hcy-lowering therapies to fight against dyslipidemia or even cardiovascular disease.