(044) A Comprehensive Meta-Analysis Assessing the Diagnostic Accuracy of Non-invasive Markers to Discriminate NAFLD and NASH

Background: Liver biopsy is a gold standard diagnostic test for liver diseases but it is an invasive and expensive procedure, however, the increasing prevalence of nonalcoholic fatty liver disease (NAFLD) has highlighted the need for accurate diagnostic tools.

Objectives: To assess the diagnostic accuracy of noninvasive markers for discrimination of NAFLD and nonalcoholic steatohepatitis (NASH).

Methods: A search of PubMed and Google Scholar yielded 83 relevant studies from January 2010 to January 2024. The studies evaluated the diagnostic accuracy of the selected noninvasive markers including the fatty liver index (FLI), hepatic steatosis index (HSI), BARD score, fibrosis-4 index (FIB-4), NAFLD fibrosis score (NFS), aspartate aminotransferase (AST) to platelet ratio index (APRI), and AST/ALT ratio, for diagnosing NAFLD. Relevant data were extracted from the selected studies, including sample size, study design, patient characteristics, and diagnostic accuracy measures for each non-invasive marker. A summary receiver operating characteristics (SROC) curve was drawn for each marker with pooled sensitivity and specificity calculated in R software. The publication bias was assessed using Deek’s test, and quality assessment, by the QUDAS-2 tool.

Results: For FLI (n=19) in diagnosing NAFLD, the Area under the curve (AUC) is 0.82 (95%CI 0.75-0.97), with a sensitivity of 0.66 (95%CI 0.61-0.81) and specificity of 0.77 (95%CI 0.70-0.83). FIB-4 (n=42) for ≥3 stages of fibrosis exhibited an AUC of 0.81 (95%CI 0.74-0.97), sensitivity of 0.58 (95%CI 0.42-0.74), and specificity of 0.83 (95%CI 0.75-0.93). NFS (n=8) score for ≥2 stages of fibrosis showed an AUC of 0.80 (95%CI 0.72-0.95), sensitivity of 0.58 (95%CI 0.34-0.80), and specificity of 0.80 (95%CI 0.75-0.93). We found that there is no publication bias reported in any markers, except the BARD score for ≥3 stages of fibrosis.

Conclusions: The FLI demonstrated robust predictive capabilities for NAFLD, where FIB-4 and NFS demonstrate better accuracy for ≥3 and ≥2 stages of fibrosis, respectively. This meta-analysis highlighted the potential of non-invasive markers in revolutionizing NAFLD diagnostics, offering a promising alternative to invasive procedures.