Metabolic Dysfunction–Associated Fibrosis 5 (MAF-5) Score Predicts Liver Fibrosis Risk and Outcome in the General Population With Metabolic Dysfunction

Laurens A. van Kleef, Sven M. Francque, Jhon E. Prieto-Ortiz, Milan J. Sonneveld, Carlos B. Sanchez-Luque, Robin G. Prieto-Ortiz, Wilhelmus J. Kwanten, Luisa Vonghia, An Verrijken, Christophe De Block, Zouhir Gadi, Harry L.A. Janssen, Robert J. de Knegt, Willem Pieter Brouwer*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

3 Citations (Scopus)
23 Downloads (Pure)


Background & Aims: There is an unmet need for noninvasive tests to improve case-finding and aid primary care professionals in referring patients at high risk of liver disease. Methods: A metabolic dysfunction–associated fibrosis (MAF-5) score was developed and externally validated in a total of 21,797 individuals with metabolic dysfunction in population-based (National Health and Nutrition Examination Survey 2017–2020, National Health and Nutrition Examination Survey III, and Rotterdam Study) and hospital-based (from Antwerp and Bogota) cohorts. Fibrosis was defined as liver stiffness ≥8.0 kPa. Diagnostic accuracy was compared with FIB-4, nonalcoholic fatty liver disease fibrosis score (NFS), LiverRisk score and steatosis-associated fibrosis estimator (SAFE). MAF-5 was externally validated with liver stiffness measurement ≥8.0 kPa, with shear-wave elastography ≥7.5 kPa, and biopsy-proven steatotic liver disease according to Metavir and Nonalcoholic Steatohepatitis Clinical Research Network scores, and was tested for prognostic performance (all-cause mortality). Results: The MAF-5 score comprised waist circumference, body mass index (calculated as kg / m2), diabetes, aspartate aminotransferase, and platelets. With this score, 60.9% was predicted at low, 14.1% at intermediate, and 24.9% at high risk of fibrosis. The observed prevalence was 3.3%, 7.9%, and 28.1%, respectively. The area under the receiver operator curve of MAF-5 (0.81) was significantly higher than FIB-4 (0.61), and outperformed the FIB-4 among young people (negative predictive value [NPV], 99%; area under the curve [AUC], 0.86 vs NPV, 94%; AUC, 0.51) and older adults (NPV, 94%; AUC, 0.75 vs NPV, 88%; AUC, 0.55). MAF-5 showed excellent performance to detect liver stiffness measurement ≥12 kPa (AUC, 0.86 training; AUC, 0.85 validation) and good performance in detecting liver stiffness and biopsy-proven liver fibrosis among the external validation cohorts. MAF-5 score >1 was associated with increased risk of all-cause mortality in (un)adjusted models (adjusted hazard ratio, 1.59; 95% CI, 1.47–1.73). Conclusions: The MAF-5 score is a validated, age-independent, inexpensive referral tool to identify individuals at high risk of liver fibrosis and all-cause mortality in primary care populations, using simple variables.

Original languageEnglish
Pages (from-to)357-367.e9
Issue number2
Publication statusE-pub ahead of print - 19 Mar 2024

Bibliographical note

Publisher Copyright: © 2024 The Authors


Dive into the research topics of 'Metabolic Dysfunction–Associated Fibrosis 5 (MAF-5) Score Predicts Liver Fibrosis Risk and Outcome in the General Population With Metabolic Dysfunction'. Together they form a unique fingerprint.

Cite this