Early serial changes in FIB-4 after out-of-hospital cardiac arrest and short-term clinical outcomes: a retrospective cohort study

The Fibrosis-4 (FIB-4) index combines age, aminotransferase levels, and platelet count. Although developed for chronic liver fibrosis, its components may reflect acute hepatic and systemic injury after cardiac arrest. The primary aim of this study was to characterize early changes in FIB-4 from admission to an approximately 72-hour repeat assessment after out-of-hospital cardiac arrest (OHCA). Secondarily, we evaluated the association of FIB-4 values with 28-day mortality and whether repeat measurement provided additional prognostic information. This single-center retrospective cohort study included adults admitted to the intensive care unit after sustained return of spontaneous circulation between May 1, 2023, and May 1, 2026. After a record-level eligibility audit, 188 patients were included; 28-day vital status was available for 182. FIB-4 was calculated from the first laboratory set obtained at intensive care unit admission and from the complete laboratory set closest to 72 h after the admission laboratory sample within a 48–96-hour window. The primary analysis assessed paired change in FIB-4. Secondary analyses examined the association of admission FIB-4 with 28-day mortality and, using a 72-hour landmark approach, the association of repeat FIB-4 with subsequent mortality through day 28 and its incremental contribution beyond the admission value. A valid repeat FIB-4 measurement was available in 122 patients. In paired analysis, median FIB-4 decreased from 5.21 at admission to 3.16 at approximately 72 h ( p < 0.001). Of the 182 patients with known 28-day vital status, 82 died (45.1%). Admission FIB-4 was higher in patients who died than in survivors (6.84 [3.25–10.75] vs. 4.14 [2.32–8.16]; p = 0.019). Each doubling of admission FIB-4 was associated with higher odds of mortality in the unadjusted analysis (OR, 1.26; 95% CI, 1.02–1.55; p = 0.029); however, in the OHCA-specific complete-case model, the adjusted association was not statistically significant (OR, 1.16; 95% CI, 0.91–1.46; p = 0.227), and adding FIB-4 to the clinical model increased the AUC only from 0.647 to 0.648 (difference, 0.001). In the 72-hour landmark cohort, 56 patients had an available pre-landmark repeat measurement and known 28-day vital status; 29 died after the landmark. Repeat FIB-4 showed weak discrimination for subsequent mortality (AUC, 0.585), and the OR per doubling was 1.23 (95% CI, 0.86–1.76; p = 0.255). Adding repeat FIB-4 to the admission value increased the AUC from 0.575 to 0.581 (difference, 0.006; 95% CI, − 0.053 to 0.205; likelihood-ratio p = 0.294). Among patients with available paired measurements, FIB-4 decreased from admission to the approximately 72-hour repeat assessment; however, the magnitude of change showed little discriminatory ability for subsequent mortality in the landmark cohort. Higher admission FIB-4 was associated with mortality in the unadjusted analysis, but this association was not statistically significant after adjustment for major OHCA-specific prognostic factors, and FIB-4 provided essentially no incremental discrimination when added to the clinical model. The incremental prognostic contribution of repeat measurement remained uncertain because of the limited landmark sample and imprecise estimates. FIB-4 should be regarded as a research marker of the acute systemic and hepatic response after cardiac arrest rather than as a stand-alone clinical prediction tool.