Fibrosis index predicts variceal bleeding and reduces need for nonselective beta-blocker in compensated cirrhosis with initial small esophageal varices without red-color sign

BACKGROUND: Various non-invasive markers predicting hepatic fibrosis are poor predictors of esophageal variceal bleeding (EVB). Elastography performs well but resource-limited. Controversy for small EV prevention also exists. We aim to investigate if a non-invasive marker could predict subsequent EVB within 1 and 2 years in patients with compensated liver cirrhosis (CLC), initial small EV without red-color sign (RCS), without use of non-selective beta-blockers (NSBB) and endoscopic variceal ligation (EVL). This marker would also be tested if it could help reduce use of NSBB, thereby avoiding potential side effects and saving medical costs. METHODS: In this retrospective cohort study, 6,803 CLC patients fulfilling the inclusion-exclusion criteria were enrolled between 2001 and 2018, and were followed-up for 1 year, 2 years. The primary outcomes were subsequent EVB within 1 and 2 years of enrollment. Another 281 CLC patients with NSBB use were compared for additional outcome analysis. RESULTS: In total, 539 patients and 710 patients experienced EVB within 1 year and 2 years, respectively. The fibrosis index (FI) with cut-off value of 3.95 showed a negative predictive value (NPV) of 94.3% and an area under receiver operating characteristic (AUROC) of 62.95% for predicting subsequent EVB within 1 year. The EVB and mortality of patients with FI <3.95 and not taking NSBB were significantly lower than those of the other 3 groups. Similar results were demonstrated within 2 years. CONCLUSIONS: In CLC patients with initial small EV and no RCS, low FI scores showed a high NPV and moderate AUROC in predicting subsequent EVB and mortalities, signifying clinically non-significant portal hypertension. Patients with low FI scores and not taking NSBB had significantly lowest EVB and mortality. The medical cost savings for cutting NSBB in these patients would be estimated at least $3 million per year in the U.S. Further randomized control trial study needed to validate this screening tool.