Comparison of non-invasive fibrosis scores to predict increased liver stiffness in the general population with unknown liver disease: Searching for the primary physician's best friend

Background: Serum-based scores (SBS) appear to be a high applicability strategy for assessment of liver fibrosis in primary care. Aim of the study was to evaluate their performance to detect ≥F2 in a general population and to design a highly-applicable strategy for screening. Methods: prospective population-based cohort study in randomly identified subjects, aged 40–70y. Medical history, blood tests and elastography were obtained, ≥F2 was determined by using LSM cutoff ≥9.2/7.8 kPa for M/XL probe and SBS diagnostic accuracies were evaluated. Results: 986 patients were analyzed. LSM prevalence estimate suggestive of ≥F2 was 1.9% and Metabolic Sindrome (MS) (OR 3.4, 1.3–9.0;p = 0.01), was the only factor independently associated with ≥F2, with increasing prevalence according to the number of criteria (0 criterion:0%,1:0.3%,2:2.8%,3:2.4%,4:6.9%,5:14.3%;p<0.001). FLI and NFS were the two best-performing tests in the cross-sectional study, with AUROCs for ≥F2 of 0.89 (95%CI,0.84- 0.95) and 0.82 (95%CI,0.74–0.90), respectively. Predefined cutoff for FLI≥60 (Sn89.5%, Sp72.1%, NPV99.9%) and NFS≥-1.455 (Sn83.3%, Sp68%, NPV99.6%) showed adequate diagnostic accuracy. Based on these findings, a 3- step algorithm strategy to detect liver fibrosis in the community setting is proposed (Sn84.2%, Sp75.2%, NVP99.6%). Conclusions: A staged risk-stratification model improves the detection of ≥F2 in the community setting, while reducing unnecessary referrals.