T-HAD Score: A Novel Diagnostic Model for Advanced Fibrosis in Nonalcoholic Fatty Liver Disease (NAFLD)

Abstract

Background and Aims. The NAFLD disease spectrum includes simple steatosis, nonalcoholic steatohepatitis (NASH), advanced fibrosis, and cirrhosis. Progression from NASH, the forerunner of developing cirrhosis, portends a poor outcome as mortality is proportionately increased. This study sought to propose a new diagnostic model for advanced fibrosis in an Asian population cohort affected with NAFLD. Methods. Cross-sectional study conducted in the Department of Medical Gastroenterology, Medical College, Trivandrum. The study period was 2 years. After excluding secondary causes of hepatic steatosis, patients were subjected to vibration-controlled transient elastography or transient elastography (VCTE or TE) to assess hepatic fibrosis. Subjects were grouped into those with advanced fibrosis (TE > 10 Kpa) and those without (TE < 10 Kpa) based on the estimation of TE. A new scoring system was then developed. This was then validated in a cohort of 84 biopsy-proven patients.Results. 1617 NAFLD patients were included in the study. Independent predictors of advanced fibrosis in this cohort were hip circumference, triglycerides, aspartate aminotransferase (AST), and diabetes mellitus (duration more than 10 years). The coefficient of beta for these variables was calculated. T-HAD score was calculated using the following formula: (Hip circumference × 0.044 + AST × 0.028 + diabetes mellitus × 3.7) − (0.03 × triglycerides). The AUROC of the T-HAD score was 0.929. The T-HAD score had a sensitivity of 90% and a specificity of 77% at a cut off of >2 for advanced fibrosis. We validated this score in another cohort of liver biopsy with advanced fibrosis. In the validation cohort, the T-HAD score had an AUROC of 0.926 in diagnosing advanced fibrosis (sensitivity of 89% and specificity of 71% at a cut off of >2). Conclusion. The T-HAD score based on data from the Asian population is a new diagnostic model which is beneficial in estimating the risk of advanced fibrosis. It is a simple yet effective tool that could be in-cooperated into day-to-day practice in a resource-limited setting.