Impact of Obesity and Alanine Aminotransferase Levels on the Diagnostic Accuracy for Advanced Liver Fibrosis of Noninvasive Tools in Patients With Nonalcoholic Fatty Liver Disease

INTRODUCTION: Some evidence suggests an interference of obesity and alanine aminotransferase (ALT) levels on the diagnostic accuracy for advanced fibrosis of noninvasive tools such as liver stiffness measurement (LSM) by FibroScan, Fibrosis-4 (FIB-4), and nonalcoholic fatty liver disease fibrosis score (NFS). We assessed whether the diagnostic accuracy of LSM, Fibrosis-4 (FIB-4), and NFS and strategies based on the combination of these tools is affected by obesity and/or ALT levels. METHODS: We analyzed data from 968 patients with a histological diagnosis of nonalcoholic fatty liver disease. FIB-4, NFS, and LSM by FibroScan were measured. RESULTS: LSM was better than both FIB-4 and NFS for staging F3-F4 fibrosis area under the receiver operating characteristic curve test (AUC) 0.863, 0.777, and 0.765, respectively; P < 0.001 for both), showing higher accuracy and higher negative predictive value (NPV), but lower positive predictive value (PPV). LSM worked less well in high ALT (>100 IU) (AUC 0.811 vs 0.877, P = 0.04; PPV 57.5% vs 62.4%; NPV 90.7% vs 94%) or obese patients (AUC 0.786 vs 0.902, P < 0.001; PPV 58.7% vs 64.8%; NPV 88.3% vs 95.2%), the latter not being affected by the M or XL probe. Consistently, LSM worked better in terms of AUC and accuracy compared with both FIB-4 and NFS only in nonobese or high ALT patients, even with always keeping a slightly lower PPV. A serial combination of FIB-4 or NFS with LSM as the second test in patients in the gray area of the first test retained—in most scenarios—similar PPV and NPV compared with LSM alone. These strategies also increased the diagnostic accuracy of about 20% in all groups of patients, even if with a lower overall accuracy in obese patients (71.3% and 67.1% for FIB-4 and NFS as the first test, respectively) compared to nonobese patients (81.9% and 82.4% for FIB-4 and NFS as the first test, respectively). CONCLUSIONS: All tested noninvasive tools have overall better NPV than PPV. LSM has a better diagnostic accuracy for advanced fibrosis than both FIB-4 and NFS only in nonobese and/or low ALT patients. Serial combination strategies are better than a single tool strategy, regardless of obesity and ALT levels, although the accuracy is lower in obese patients.