Comparison of laboratory tests, ultrasound, or magnetic resonance elastography to detect fibrosis in patients with nonalcoholic fatty liver disease: A meta‐analysis

Many noninvasive methods for diagnosing liver fibrosis (LF) have been proposed. To determine the best method for diagnosing LF in nonalcoholic fatty liver disease (NAFLD), we conducted a systemic review and meta-analysis to compare the performance of aspartate aminotransferase to platelets ratio index (APRI), fibrosis-4 index (FIB-4), BARD score, NAFLD fibrosis score (NFS), FibroScan, shear wave elastography (SWE), and magnetic resonance elastography (MRE) for diagnosing LF in NAFLD. We compared the sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve (AUROC) of these noninvasive methods for detecting significant fibrosis (SF), advanced fibrosis (AF), and cirrhosis. Heterogeneity was explored using meta-regression. Sixty-four articles with a total of 13,046 NAFLD subjects were included. The overall mean prevalence of SF, AF, and cirrhosis was 45.0%, 24.0%, and 9.4% in NAFLD patients, respectively. With an APRI threshold of 1.0 and 1.5, the sensitivities and specificities were 50.0% and 84.0% and 18.3% and 96.1%, respectively, for AF. With a FIB-4 threshold of 2.67 and 3.25, the sensitivities and specificities were 26.6% and 96.5% and 31.8% and 96.0%, respectively, for AF. The summary sensitivities and specificities of BARD score (threshold of 2), NFS (threshold of -1.455), FibroScan M (threshold of 8.7-9), SWE, and MRE for detecting AF were 0.76 and 0.61, 0.72 and 0.70, 0.87 and 0.79, 0.90 and 0.93, and 0.84 and 0.90, respectively. The summary AUROC values using APRI, FIB-4, BARD score, NFS, FibroScan M probe, XL probe, SWE, and MRE for diagnosing AF were 0.77, 0.84, 0.76, 0.84, 0.88, 0.85, 0.95, and 0.96, respectively.MRE and SWE may have the highest diagnostic accuracy for staging fibrosis in NAFLD patients. Among the four noninvasive simple indexes, NFS and FIB-4 probably offer the best diagnostic performance for detecting AF. (Hepatology 2017;66:1486-1501).