Prevalence of Fibrosis and Applicability of Lab-Based Noninvasive Tests From Primary to Tertiary Care

Knowing the prevalence of fibrosis, the accuracy of noninvasive tests, and their applicability in referral strategies of individuals screened or surveilled for steatotic liver disease is crucial for individual and policy decision making.

Five cohorts spanning primary to tertiary care between 2016 and 2024 were retrospectively included: individuals from the general population (colonoscopy screening, n = 1185, cohort I; primary care); at-risk individuals with metabolic dysfunction (overweight n = 1333, cohort II; prediabetes, n = 996, cohort III; secondary care); and subjects referred to hepatology clinics (n = 2598/n = 2397, cohort IV/V; tertiary care).

Mean controlled attenuation parameter was 262 to 294 dB/m, and 41% to 64% showed hepatic steatosis. Suspected fibrosis (liver stiffness measurement [LSM] ≥8 kPa) was found in 6.4% of cohort I, 7.3%/6.8% of cohort II/III, and 21%/30% of cohort IV/V, whereas advanced fibrosis (≥12 kPa) was rare (1.2% in cohort I, 1.4%/0.8% in cohorts II/III, and 12%/17% in cohorts IV/V, respectively). The prevalence of fibrosis was especially high in subjects with obesity or diabetes, but not in overweight or prediabetes. Sensitivity of Fibrosis-4 index ≥1.3 for liver stiffness measurement ≥8 kPa was low in primary and secondary care (33%-44%) with simultaneously low positive predictive value (8.3%-12%), and overall low accuracy (area under the precision recall curve [AUPRC], 0.09-0.15). LiverPRO (AUPRC, 0.12-0.24) and LiverRisk score (AUPRC, 0.13-0.24) were numerically more accurate, with LiverPRO being the most sensitive and LiverRisk score the most specific first-line test. In tertiary care, LiverPRO was comparatively accurate, but more sensitive than Fibrosis-4 index.

Prevalence of fibrosis in contemporary patients with steatotic liver disease was ∼7% in primary/secondary care, being especially pronounced in individuals with diabetes or obesity. LiverPRO and LiverRisk score optimize referral pathways.