RESUMEN
OBJECTIVES: Liver fibrosis is one of the most important predictors of mortality related to nonalcoholic fatty liver disease (NAFLD). The use of noninvasive markers has the advantage of a simple and low-cost evaluation. The aim of this study was to evaluate the performance of six noninvasive scores for the diagnosis of advanced liver fibrosis in morbidly obese patients. MATERIAL AND METHODS: A retrospective study validation included 323 morbidly obese patients undergoing bariatric surgery. Advance fibrosis was defined as stage 3 and 4 (septal fibrosis or cirrhosis). Accuracy, sensitivity, specificity, positive (PPV) or negative (NPV) predictive value, and positive (PLR) or negative (NLR) likelihood ratio test of the following noninvasive liver fibrosis scores were evaluated: aspartate aminotransferase (AST)/alanine aminotransferase (ALT) ratio (AAR); AST to platelet ratio index (APRI); BARD; FIB4, NAFLD fibrosis score (NFS) and BAAT, which were compared with the histological findings of the intraoperative liver biopsy. The cutoff points established in the validation studies were used: AAR > 1; APRL > 0.98; BARD ≥ 2; FIB4 > 2.67; NFS > 0.676 and BAAT > 1. RESULTS: Twenty-nine patients (8.97%) presented advanced fibrosis. APRI presented the higher specificity (99.61%), PPV (85.71%), PLR (62.5) and accuracy (0.93). FIB4 was the second test in accuracy (0.9) and in PLR (10.53). BAAT presented the highest sensitivity (73.08%) and NPV (94.78%); NFS the lowest sensitivity (12,5%), and BARD the lowest accuracy (0.44). CONCLUSION: APRI and FIB-4 were the tests with best performance to predict advanced fibrosis.