Validation of magnetic resonance elastography plus fibrosis-4 for significant fibrosis in nonalcoholic fatty liver disease

Background and Aim MEFIB (the combination of magnetic resonance elastography [MRE] >= 3.3 kPa and fibrosis-4 (FIB-4) >= 1.6) is useful for detecting patients with significant fibrosis (fibrosis stage >= 2) having nonalcoholic fatty liver disease (NAFLD). However, age-dependent thresholds of FIB-4 have been proposed, and it remains unclear whether MEFIB could be applied with the same FIB-4 threshold in a different cohort. Therefore, in this study, we examined the best threshold of FIB-4 and validated the utility of MEFIB. Methods This study included 105 biopsy-proven NAFLD patients with contemporaneous MRE assessment. The primary outcome was a diagnostic accuracy for significant fibrosis. Results The median (interquartile range) age was 65 (58-72) years, and significant fibrosis was 76.2% (80/105). FIB-4 of 2.1 was defined as the best threshold for significant fibrosis in the cohort. The area under the receiver operating characteristics curves (AUROCs) of the combination of MRE and FIB-4 (MRE >= 3.3 kPa + FIB-4 >= 1.6: 0.80, MRE >= 3.3 kPa + FIB-4 >= 2.1: 0.84) were higher than those of each index alone (MRE >= 3.3 kPa: 0.76, FIB-4 >= 1.6: 0.72, and FIB-4 >= 2.1: 0.77), but AUROCs of MRE >= 3.3 kPa + FIB-4 >= 1.6 and MRE >= 3.3 kPa + FIB-4 >= 2.1 were equivalent (P = 0.3). Conclusions MEFIB is useful for detecting patients with significant fibrosis and could be utilized in a different cohort without changing the threshold of FIB-4, and it may then be used as a two-step screening strategy.