Serial FIB-4 variability and persistent elevation were associated with incident heart failure in type 2 diabetes independently of chronological age, but adding these serial features gave only a small, statistically uncertain improvement in prediction, so external validation is needed before clinical use.
Greater within-person FIB-4 variability was associated with a higher hazard of incident heart failure in patients with type 2 diabetes.
Persistently high FIB-4 was associated with a higher hazard of heart failure compared with a never-high pattern.
Worsening to a high FIB-4 was associated with a modestly higher heart failure hazard compared with never-high.
Improving from a high FIB-4 was not associated with higher heart failure hazard compared with never-high.
Adding serial FIB-4 features to a prediction model produced only a small, statistically uncertain improvement in discrimination in temporal validation.
The study used a 24-month landmark design with a lag period to reduce bias from reverse causation, using routinely collected clinical data from 2000 to 2025.
The authors conclude that serial FIB-4 variability and persistent elevation are associated with heart failure independently of age and may reflect longitudinal multisystem cardiometabolic burden, but require external validation before clinical application.
This study looked at a blood-test-based score called FIB-4, usually used to estimate liver scarring, in over 8,000 people with type 2 diabetes. The researchers asked whether how much a person's FIB-4 score fluctuated over about 18 months, and whether it stayed high, was related to later development of heart failure. Over roughly 50,000 person-years of follow-up, 1,811 people developed heart failure. People whose FIB-4 scores varied more had a higher risk of heart failure, and so did those whose scores stayed persistently high (about 73% higher hazard than people never high). Those whose scores rose to high had a smaller increased hazard (about 28%). People whose scores came down from high had no clear increase compared with those who were never high. These results held after accounting for age, which is part of the FIB-4 calculation. However, adding these repeated FIB-4 measures to a prediction model improved its accuracy only slightly, and the improvement was statistically uncertain. This research suggests that changes in FIB-4 over time may reflect broader cardiometabolic health, but it does not show they are useful for predicting individual risk. The findings come from routine clinical records and need to be confirmed in other populations before any clinical use.
Nagy A, Ghanem A, Ulambayar B, Móré M, Harangi M. (2026). Serial fibrosis-4 variability and longitudinal patterns are associated with incident heart failure in type 2 diabetes: a 24-month landmark cohort study.. Frontiers in endocrinology. https://doi.org/10.3389/fendo.2026.1963810