Triglycerides to high‐density lipoprotein cholesterol ratio is an independent predictor of incident fatty liver; a population‐based cohort study

Abstract Background Triglycerides (TG) to high‐density lipoprotein cholesterol (HDL‐C) ratio (TG/HDL‐C) has been recommended for surrogates of insulin resistance. However, it remains to be elucidated the association between TG/HDL‐C and incident fatty liver. Aims To investigate the association between TG/HDL‐C and incident fatty liver. Methods We performed population‐based historical cohort study consisted with 4518 healthy Japanese who received yearly health‐checkup programmes over decade. Fatty liver was diagnosed using ultrasonography. Results During the observation periods, 38.8% (case/N = 1023/2637) of men and 17.2% (case/N = 324/1881) of women developed fatty liver. Adjusting odds ratio of TG/HDL‐C for incident fatty liver were 1.59 (95% confidence interval (CI) 1.42–1.79, P < 0.0001) in men and 2.50 (95% CI 1.80–3.51, P < 0.0001) in women. In addition, adjusting odds ratio of TG/HDL‐C for incident non‐alcoholic fatty liver disease were 1.55 (95% CI 1.35–1.77, P < 0.0001) in men and 2.72 (95% CI 1.88–3.95, P < 0.0001) in women. According to the receiver operator characteristic (ROC) analysis, the optimal cut‐off point of TG/HDL‐C for incident fatty liver was 0.88 (area under the ROC curve (AUC) 0.67 [95% CI 0.65–0.69], sensitivity = 0.64, specificity = 0.60, P < 0.0001) in men and 0.64 (AUC 0.69 [95% CI 0.66–0.72], sensitivity = 0.50, specificity = 0.78, P < 0.0001) in women. Conclusions The TG/HDL‐C could predict the incident fatty liver. Thus, it is important to check TG/HDL‐C and lifestyles modification is needed for preventing future fatty liver disease in patients with high TG/HDL‐C.