Non-invasive score identifies ultrasonography-diagnosed non-alcoholic fatty liver disease and predicts mortality in the USA

BACKGROUND: Several non-invasive prediction scores for non-alcoholic fatty liver disease (NAFLD) have been developed, but their performance has not been compared and validated in the same population, and whether these prediction scores can predict clinical outcomes remains unknown. In this study, we aimed to validate and compare the performance of four NAFLD prediction scores: fatty liver index, hepatic steatosis index, lipid accumulation product, and NAFLD liver fat score (LFS), and to evaluate the ability of the best NAFLD prediction score to predict mortality. METHODS: We analyzed data from the National Health and Nutrition Examination Survey conducted in 1988 to 1994, and subsequent follow-up data for mortality up to December 31, 2006. NAFLD was defined by ultrasonographic detection of hepatic steatosis in the absence of other known liver diseases. RESULTS: In a group of 5,184 participants, LFS consistently showed the highest area under the curve for predicting the presence of NAFLD. During a median follow-up of 14.7 years (range 0.1 to 18.2 years) and 83,830.5 person-years, participants in the high LFS group (LFS ≥1.257) had a higher cardiovascular and liver-related mortality than participants in the low (LFS ≤ −1.413; cardiovascular hazard ratio (HR) = 2.24, 95% CI 1.03 to 4.88; liver HR = 31.25, 95% CI 3.13 to 333.33) or intermediate (−1.413 < LFS < 1.257; cardiovascular HR = 2.3, 95% CI 1.19 to 4.48; liver HR = 30.3, 95% CI 4 to 250) LFS groups in the fully adjusted model. Similar results were obtained when LFS was treated as a continuous variable. CONCLUSIONS: LFS is the best non-invasive prediction score for NAFLD, and people with a high LFS score have an increased risk for cardiovascular and liver-related mortality. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12916-014-0154-x) contains supplementary material, which is available to authorized users.