RESUMO
BACKGROUND & AIMS: There is limited knowledge regarding the longitudinal utility of biomarkers of fibrosis, such as the nonalcoholic fatty liver disease (NAFLD) fibrosis score (NFS) or the fibrosis-4 score (FIB-4) score. We examined longitudinal changes in the NFS and the FIB-4 score in patients with NAFLD, with and without clinically significant fibrosis (CSF). METHODS: We performed a retrospective study of 230 patients with NAFLD, collecting clinical and laboratory records to calculate NFS and FIB-4 scores at 6 monthly intervals for 5 years before hepatology assessment of fibrosis. Linear mixed models with random intercept and slope and adjusted for age at baseline were used to assess the progression of NFS and log-transformed FIB-4 scores over time in subjects with and without CSF, determined by liver stiffness measurements of 8.2 kPa or greater. RESULTS: Patients had a median of 11 (minimum, 10; maximum, 11) retrospective observations over a median time period of 5 years (minimum, 4.5 y; maximum, 5 y). Of patients with low baseline NFS and FIB-4 scores, 31.11% and 37.76%, respectively, had CSF at the time of hepatology assessment. There was a correlation between NFS and log10 FIB-4 over time (repeated measure r = 0.55; 95% CI, 0.52-0.59). The rate of increase in NFS and log10 FIB-4 was significantly higher in patients with than without CSF (both P < .001). Predicted NFS increased by 0.17 and 0.06 units per year in subjects with and without CSF, respectively. Predicted log10 FIB-4 score increased by 0.032 and 0.0003 units per year in subjects with and without CSF, respectively. CONCLUSIONS: Noninvasively measured fibrosis scores increase progressively in patients with NAFLD and CSF. Further studies are needed to determine whether repeated measurements can identify patients at risk for CSF.