Estimation of visceral fat is useful for the diagnosis of significant fibrosis in patients with non-alcoholic fatty liver disease

Abstract

Background: Obesity is a risk factor for non-alcoholic fatty liver disease (NAFLD), although obese patients with NAFLD do not always develop significant fibrosis. The distribution of body fat could predict the risk of NAFLD progression.

Aim: To investigate the role of bioelectrical impedance-estimated visceral fat (VF) in assessing NAFLD severity.

Methods: In this cross-sectional study, patients with biopsy-proven NAFLD were prospectively included. All patients underwent anthropometric evaluation, blood tests and bioelectrical impedance analysis.

Results: Between 2017 and 2020, 119 patients were included [66.4% male, 56 years (SD 10.7), 62.2% obese, 61.3% with metabolic syndrome]. Sixty of them (50.4%) showed significant fibrosis (≥ F2) in liver biopsy. Age, VF and metabolic syndrome were associated with significant fibrosis (61 years vs 52 years, 16.4 vs 13.1, 73.3% vs 49.2%, respectively; P < 0.001 for all). In the multivariate analysis, VF and age were independently associated with significant fibrosis (VF, OR: 1.11, 95%CI: 1.02-1.22, P = 0.02; age, OR: 1.08, 95%CI: 1.03-1.12, P < 0.01). A model including these variables showed and area under the receiver operating characteristic curve (AUROC) of 0.75, which was not inferior to transient elastography or NAFLD fibrosis score AUROCs. We developed a nomogram including age and VF for assessing significant fibrosis in routine practice.

Conclusion: VF is a surrogate marker of liver fibrosis in patients with NAFLD. Bioelectrical impedance analysis is an inexpensive and simple method that can be combined with age to guide patient referral when other resources may be unavailable.

Keywords: Bioimpedanciometry; Liver fibrosis; Metabolic syndrome; Non-alcoholic fatty liver disease; Obesity; Visceral fat.

Figure 1

Visceral fat measurement by bioimpedanciometry, according to histological fibrosis stage. A: Visceral fat measurements increased along with fibrosis stage assessed by histological analysis (F0-1, 12; F2-3, 14; F4, 16; Kruskal-Wallis ^cP < 0.001). A line can be fit by linear regression, showing linear association (r² = 0.11, ^cP < 0.001); B: Visceral fat measurements were greater for those patients with significant fibrosis (16.3 vs 13.1, ^cP < 0.001).

Figure 2

Area under the receiver operating characteristic curve. A: Receiver operating characteristic (ROC) curve for non-invasive diagnosis of significant liver fibrosis by a model including age and visceral fat; B: Comparison of the areas under ROC curves for a model using age and visceral fat versus liver elastography measurement, to predict significant liver fibrosis. Circles denote our model, triangles indicate non-alcoholic fatty liver disease fibrosis score and crosses denote liver elastography.

Figure 3

Nomogram for assessing the probability of significant liver fibrosis in a clinically useful manner. With the variables resulting from the multivariate regression model, we built an easy-to-use visual tool. In an individual patient, visceral fat levels and age correspond to a score. Combining these scores gives a total score that can be converted to a probability of that patient having significant fibrosis in liver biopsy. For example, a patient with a visceral fat level of 12 (score 2) and with 55 years old (score 7) would have a total score of 9 and a corresponding probability of histological significant fibrosis of 43%.