Assessment of asymptomatic liver fibrosis in alcoholic patients using fibroscan: prospective comparison with seven non-invasive laboratory tests

Background Systematic screening for liver fibrosis in heavy-drinking patients is a challenge. Aims To assess Fibroscan for non-invasive diagnosis of asymptomatic liver fibrosis in alcohol abuse patients, to determine diagnostic liver stiffness cut-off values and to compare performance of Fibroscan with seven non-invasive laboratory tests. Methods One hundred and three alcoholic patients were studied. Liver fibrosis was staged by METAVIR system. Fibroscan, Fibrotest, Fibrometer, Hepascore, APRI, PGA, PGAA and hyaluronic acid tests were performed. Liver stiffness cut-offs were determined using receiver-operating characteristic (ROC) curves. Results Liver stiffness was correlated with fibrosis (r = 0.72, P < 0.014), with median at 5.7, 6.3, 8.4, 15 and 47.3 kPa for F0 (n = 8), F1 (n = 18), F2 (n = 24), F3 (n = 20) and F4 (n = 33) stage fibrosis respectively. For Fibroscan, areas under ROC curves (AUROCs) were 0.84 (95% CI: 0.73-0.95) (F >= 1), 0.91 (0.85-0.98) (F >= 2), 0.90 (0.82-0.97) (F >= 3) and 0.92 (0.87-0.98) (F = 4), yielding diagnostic stiffness cut-offs of 5.9 (F >= 1), 7.8 (F >= 2), 11 (F >= 3) and 19.5 (F4) kPa. Sensitivity, specificity, PPV and NPV were 80%, 90.5%, 93% and 70% for F >= 2, and 85.7%, 84.2%, 68.6% and 87.9% for F = 4. Performance of Fibroscan was higher than seven laboratory tests, for which AUROCs ranged from 0.66 to 0.77 (F >= 1), from 0.54 to 0.82 (F >= 2), from 0.43 to 0.88 (F >= 3) and from 0.56 to 0.89 (F = 4), with significant difference only vs. APRI (P < 0.001) and Hepascore (P = 0.04). Combining Fibroscan with each tests did not improve performance. Conclusions Fibroscan is effective to assess liver fibrosis in alcoholic patients. Instant screening of liver fibrosis in heavy drinkers is feasible without liver biopsy.