Intrahepatic fat, abdominal adipose tissues, and metabolic state: magnetic resonance imaging study.

BACKGROUND

Intrahepatic fat (IHF) is best known to associate with waist circumference (WC) and visceral adipose tissue (VAT), but its relation to abdominal subcutaneous adipose tissue is controversial. While IHF ≥ 5% dichotomously defines fatty liver, %IHF is rarely considered as a continuous variable that includes the normal range. In this study, we aimed to evaluate %IHF association with abdominal fat subdepots, pancreatic, and renal-sinus fats.

METHODS

We evaluated %IHF, abdominal fat subdepots, %pancreatic, and renal-sinus fats, among individuals with moderate abdominal obesity, using 3-Tesla magnetic resonance imaging.

RESULTS

Among 275 participants, %IHF widely ranged (0.01%-50.4%) and was lower in women (1.6%) than men (7.3%; P < .001). In an age, sex, and WC-adjusted models, VAT area (P < .006) was directly associated with %IHF, while superficial-subcutaneous adipose tissue proportion was inversely associated with %IHF (P < .006). In these models, renal-sinus fat was positively associated with %IHF (P = .005). In an age, sex, WC, and VAT-adjusted models, elevated liver enzymes, glycemic, lipid, and inflammatory biomarkers were associated with increased %IHF (P < .003 for all). In these models, the associations remained robust even within the normal range strata of IHF < 5% for triglycerides and chemerin (P ≤ .004 for all). For the diagnosis of fatty liver, the joint area under the curve of WC, alanine-aminotransferase, triglycerides/high-density lipoprotein cholesterol, and homeostasis model assessment of insulin resistance was 0.84(95% CI, 0.79-0.89).

CONCLUSIONS

Intrahepatic fat is differentially associated with abdominal fat subdepots. Intrahepatic-fat as a continuous variable could be predicted by specific traditional parameters, even within the current normal range, and partially independent of VAT.