Poor accuracy and sustainability of the first-step FIB4 EASL pathway for stratifying steatotic liver disease risk in the general population

Background and AimsThe European Association for the Study of the Liver introduced a clinical pathway (EASL CP) for screening significant/advanced fibrosis in people at risk of steatotic liver disease (SLD). We assessed the performance of the first-step FIB4 EASL CP in the general population across different SLD risk groups (MASLD, Met-ALD and ALD) and various age classes. MethodsWe analysed a total of 3372 individuals at risk of SLD from the 2017-2018 National Health and Nutrition Examination Survey (NHANES17-18), projected to 152.3 million U.S. adults, 300,329 from the UK Biobank (UKBB) and 57,644 from the Biobank Japan (BBJ). We assessed liver stiffness measurement (LSM) >= 8 kPa and liver-related events occurring within 3 and 10 years (3/10 year-LREs) as outcomes. We defined MASLD, MetALD, and ALD according to recent international recommendations. ResultsFIB4 sensitivity for LSM >= 8 kPa was low (27.7%), but it ranged approximately 80%-90% for 3-year LREs. Using FIB4, 22%-57% of subjects across the three cohorts were identified as candidates for vibration-controlled transient elastography (VCTE), which was mostly avoidable (positive predictive value of FIB4 >= 1.3 for LSM >= 8 kPa ranging 9.5%-13% across different SLD categories). Sensitivity for LSM >= 8 kPa and LREs increased with increasing alcohol intake (ALD>MetALD>MASLD) and age classes. For individuals aged >= 65 years, using the recommended age-adjusted FIB4 cut-off (>= 2) substantially reduced sensitivity for LSM >= 8 kPa and LREs. ConclusionsThe first-step FIB4 EASL CP is poorly accurate and feasible for individuals at risk of SLD in the general population. It is crucial to enhance the screening strategy with a first-step approach able to reduce unnecessary VCTEs and optimise their yield.