Fatty Liver Index Shows Poor Discrimination for <scp>MRI</scp> ‐Measured Hepatic Steatosis in African Compared with European Adults

Abstract

ABSTRACT Background and Aim The Fatty Liver Index (FLI) is a proxy for hepatic steatosis, but its performance against magnetic resonance imaging (MRI) is unknown in African populations. We evaluated the discriminatory performance of the FLI for detecting hepatic steatosis in European men (EM) and West African men (WAM) in the UK, and Black South African women (BSAW). Methods This cross‐sectional study included 57 EM and 55 WAM from the UK, and 45 BSAW. MRI‐derived hepatic steatosis was defined as liver fat ≥ 5%. Discriminatory performance of FLI was assessed using correlation analyses and receiver operating characteristic area under the curves (AUROC). Results FLI correlated strongly with MRI‐measured liver fat and showed good discrimination for hepatic steatosis in EM (r s = 0.74, p &lt; 0.0001; AUROC: 0.89 [0.81–0.98]), but weakly or not at all in WAM (r s = 0.31, p = 0.030; AUROC: 0.65 [0.46–0.85]) and BSAW (r s = 0.20, p = 0.17; AUROC: 0.57 [0.40–0.75]). Within the UK cohort, FLI performed significantly better in EM than in WAM ( p = 0.027). Conclusion FLI reliably detects hepatic steatosis in Europeans but performs poorly in Africans when validated against MRI, highlighting the need for population‐specific diagnostic tools.