FIB-4 Outperforms NFS for Detecting Advanced Fibrosis in Lean NAFLD


A new study revealed differences in the diagnostic performance of the FIB-4 and NFS, indicating that conventional NFS cutoff values may be unsuitable for screening lean NAFLD patients.

Mindie Nguyen, MD, MAS | Credit: Stanford Health Care

Mindie Nguyen, MD, MAS

Credit: Stanford Health Care

A recent study of individuals with biopsy-proven nonalcoholic fatty liver disease (NAFLD) in Asia found notable differences in the diagnostic performance of the fibrosis-4 index (FIB-4) and NAFLD fibrosis score (NFS) based on patients’ BMI.

“To our knowledge, our study stands out as the first to compare the diagnostic performance of the FIB-4 and NFS in a large cohort of lean individuals with liver biopsy-confirmed NAFLD,” wrote investigators.

The trial was conducted using anonymous clinical data from 1995 to 2019 and included over 1500 patients from 6 centers in Taiwan, Korea, and Japan. Findings suggest the screening tools did not differ significantly in their ability to diagnose advanced fibrosis, but the sensitivity of NFS at the current conventional cutoff value of 23 kg/m2 was an inadequate tool for lean NAFLD patients compared to FIB-4. NFS sensitivity and specificity showed increasing and decreasing tendency according to the BMI quartiles (P < .001), but sensitivity and specificity for the FIB-4 did not (P = 0.5, P = .20). Areas under the operating characteristic curve of the FIB-4 and NFS were similar in the lean group (0.807 vs 0.790; P = .09), but sensitivity of the current NFS cutoff values in the lean group were notably lower than observed in the FIB-4 (54.4% vs 81.8%; P = .03).

Investigators sought to develop a more comprehensive understanding of diagnostic performance and adequate cutoffs for FIB-4 and NFS, especially in lean individuals with NAFLD. Despite being the recommended first step for stratifying liver fibrosis risk in these patients by the American Gastroenterological Association (AGA) Institute’s Clinical Practice Updates Committee and Governing Board, diagnostic performance and adequate cutoffs for the FIB-4 and NFS in lean individuals had not been evaluated. FIB-4 and NFS are commonly recommended as the initial step for identifying high-risk groups among obese NAFLD patients. However, lean NAFLD patients account for roughly 15% to 19% of individuals with NAFLD.

In this retrospective study, investigators designed their analyses to compare the performance of the FIB-4 and NFS between lean and nonlean groups using anonymous clinical data of over 1500 patients with biopsy-proven NAFLD from 6 referral centers across Taiwan, Korea, and Japan. The overall study cohort had a mean age of 46.1 (Standard Deviation [SD], 16.4) years and 788 (52.5%) were male.

Initial analysis indicated lean NAFLD patients comprised 7.7% of the cohort (n = 115) and had a much lower mean BMI than the nonlean patients (21.3 [1.6] vs 30.6 [6.4]; P < .001). Investigators also noted the prevalence of advanced hepatic fibrosis was similar between the lean and nonlean NAFLD groups (20.0% vs 20.9%; P = .82).

A second cohort was created using propensity score matching using age and sex as covariates to offset background differences between the groups, which produced a matched cohort of 114 lean and 452 nonlean NAFLD patients, 233 of whom were male (41.2%) with a mean (SD) age of 52.3 (15.1) years. Again, the mean (SD) BMI of lean NAFLD patients was notably lower than the non-lean patients (21.2 [1.6] vs 29.0 [4.7]; P < .001). Prevalence of advanced hepatic fibrosis in nonlean patients was higher than lean patients (26.8% vs 19.3%).

No significant differences in the sensitivity or specificity were observed in the age- and sex-matched cohort (81.8% vs 83.5%; P = .85 and 52.2% vs 52.0%; P = .97, respectively) of FIB-4 between the lean and nonlean groups. The NFS, however, had a lower sensitivity (54.5% vs 72.7%; P = .04) and higher specificity (76.1% vs 56.8%; P < .001) in lean patients compared to nonlean patients. Furthermore, the NFS sensitivity in lean patients was lower than that of the FIB-4 (54.5% vs 81.8%; P = .03). Overall, the sensitivity of the NFS increased (53.6% to 85.7%; P for trend < .001) while the specificity decreased (73.7% to 51.5%; P for trend < .001) as BMI increased.

“It will be necessary to lower the cutoff value of the NFS to increase its sensitivity to be comparable with that in nonlean patients or that of the FIB-4. However, the use of different cutoff values according to the BMI condition represents a significant hurdle to the clinical utilization of NFS,” noted investigators.


Park H, Yoon EM, Ito T, et al. Diagnostic Performance of the Fibrosis-4 Index and Nonalcoholic Fatty Liver Disease Fibrosis Score in Lean Adults With Nonalcoholic Fatty Liver Disease. Jama Network Open. Published online August 17, 2023. doi:10.1001/jamanetworkopen.2023.29568

Related Videos
Video 3 -  4 KOLs are featured in, "Collaborating with Specialists: Primary Care Clinician’s Evolving Role in HE Management"
Video 3 -  4 KOLs are featured in, "HE Experts on Symptoms, Risk Factors, Diagnosis Challenges, and Severity Grading"
© 2024 MJH Life Sciences

All rights reserved.