Prognostic utility of Fibrosis-4 Index for risk of subsequent liver and cardiovascular events, and all-cause mortality in individuals with obesity and/or type 2 diabetes: a longitudinal cohort study

Abstract

Background: The Fibrosis-4 Index (FIB-4) is used as a non-invasive tool for the presence of advanced liver fibrosis in metabolic dysfunction-associated steatotic liver disease and type 2 diabetes. However, evidence for an association between FIB-4 and risk of mortality and/or liver-related clinical outcomes is limited. The aim of this study was to investigate the association between FIB-4 and subsequent liver events, cardiovascular events, and all-cause mortality in individuals with obesity and/or type 2 diabetes examined in routine general practice.

Methods: This was a longitudinal cohort study in which eligible adults had obesity and/or type 2 diabetes and ≥1 FIB-4 score calculable from UK Clinical Practice Research Datalink GOLD after 1 January 2001. No alcohol-related disorders and/or chronic liver diseases (except non-alcoholic fatty liver disease) and/or no prescriptions of drugs inducing liver disease were permitted. Individuals were followed until time of first event, 10 years, or 1 January 2020. Analyses were conducted using Aalen-Johansen cumulative incidence functions and Cox proportional hazards models.

Findings: Among 44,481 included individuals (mean age 58·8 years; 54% female), there were 979 liver, 6002 cardiovascular, and 8971 mortality events during the 10 years of follow-up. At 10 years, the cumulative incidence of liver events in the high (>2·67), indeterminate (1·30-2·67), and low (<1·30) baseline FIB-4 risk groups were 15%, 3%, and 1%, respectively. Age- and sex-adjusted hazard ratios (HRs) for liver events were elevated in high (16·46; 95% confidence interval [CI] 13·65-19·85) and indeterminate (2·45; 95% CI 2·07-2·90) versus low FIB-4 risk groups. Similar results were found for cardiovascular events and all-cause mortality. Among 20,433 individuals with ≥2 FIB-4 measurements, increase/decrease in FIB-4 12 months after baseline was directly associated with risk of liver events: compared with individuals with low baseline FIB-4 and no change in FIB-4 (reference), the adjusted HR (95% CI) for those with high baseline FIB-4 was 24·27 (16·98-34·68) with a one-unit FIB-4 increase, and 10·90 (7·90-15·05) with a one-unit decrease.

Interpretation: In addition to its value as a diagnostic tool, FIB-4 has clinical utility as a prognostic biomarker. Sequential measurement provides a pragmatic, tractable monitoring biomarker that refines risk assessment for liver events, cardiovascular events, and mortality.

Funding: Novo Nordisk A/S.

Keywords: Cardiovascular events; FIB-4; Hepatic decompensation; Liver events; Metabolic dysfunction-associated steatotic liver disease.

Fig. 1

Flowchart. CPRD = Clinical Practice Research Datalink. E = events. FIB-4 = Fibrosis-4 Index. HES = Hospital Episode Statistics. m = month. ONS = Office of National Statistics. ^aIndividuals with no follow-up time (liver: n = 7; cardiovascular: n = 18; mortality: n = 4), defined as individuals with an event occurring on the date of FIB-4 measurement (ie, baseline) were excluded from analyses.

Fig. 2

Cumulative incidence according to baseline FIB-4 for A) liver events, B) cardiovascular events, C) all-cause mortality. Event risks plotted as Aalen-Johansen cumulative incidence functions, with all-cause mortality included as a competing risk factor in plots of liver and cardiovascular events. FIB-4 risk categories: low <1·30; indeterminate 1·30–2·67; high >2·67. FIB-4 = Fibrosis-4 Index.

Fig. 3

Cumulative incidence of liver events according to a 12-month increase or decrease in FIB-4 by baseline FIB-4 category. Event risks plotted as Aalen-Johansen cumulative incidence functions, with all-cause mortality included as a competing risk factor. FIB-4 risk categories: low < 1·30; indeterminate 1·30–2·67; high >2·67. FIB-4 = Fibrosis-4 Index.

Fig. 4

Hazard ratios of liver events for 12-month changes in FIB-4. Reference is patients with low baseline FIB-4 and no change in FIB-4 (HR = 1). Time since FIB-4 measurement as the underlying timescale and age included in strata (baseline hazard). The model included: change in FIB-4 (continuous), baseline FIB-4 (categorical), sex (categorical), and the interaction change in FIB-4 (continuous)∗baseline FIB-4 (categorical). FIB-4 risk categories: low <1·30; indeterminate 1·30–2·67; high >2·67. CI = confidence interval. FIB-4 = Fibrosis-4 Index. HR = hazard ratio.