Correlation Between Transient Elastography and Non-invasive Biomarker Scores for the Detection of Liver Fibrosis

Background and aim The study aimed to address the need for reliable and non-invasive biomarkers (NIBM) for detecting fibrosis among patients with chronic liver disease (CLD). Material and methods This was a diagnostic validation study executed at the Department of Gastroenterology, Jinnah Hospital, Lahore. The study was carried out from July 2023 to June 2024, enrolling a total of 88 patients using non-probability consecutive sampling. Patients with chronic liver disease (CLD) due to chronic viral hepatitis B (HBV), hepatitis C (HCV), and non-alcoholic fatty liver disease (NAFLD) were included in the study. A cut-off value of 12.5 kPa was used to label fibrosis using transient elastography. Blood samples were collected for recording values of alanine aminotransferase (ALT), aspartate aminotransferase (AST), and platelet count, and the appropriate formulas were applied to calculate the aspartate aminotransferase to platelet ratio index (APRI), Fibrosis-4 (FIB-4), AST/ALT-to-platelet ratio index (AARPRI), and BARD scores (a scoring system for NAFLD fibrosis that predicts the risk of advanced fibrosis in NAFLD patients; the components include body mass index (BMI), aspartate aminotransferase (AST)/alanine aminotransferase (ALT) ratio, and diabetes status). Spearman's rank correlation coefficient was used to assess the strength and direction of the association between these variables. Results Among the 88 patients, 61.4% were male and HCV was the most common cause of CLD (63.6%). Statistically significant correlations were found between transient elastography and NIBM: FIB-4 (p < 0.001, area under receiver operating characteristic curves (AUROC), 0.835; sensitivity, 47.7%; specificity, 90.9%), APRI (p = 0.020; AUROC, 0.769; sensitivity, 25.0%; specificity, 93.2%), AARPRI (p = 0.011, AUROC, 0.782), and BARD (p = 0.033; AUROC, 0.648). FIB-4 demonstrated the strongest correlation with liver stiffness measurements (LSM) (r = 0.617), indicating its reliability in detecting liver fibrosis, whereas the aspartate aminotransferase to alanine aminotransferase ratio (AAR) showed the weakest correlation (r = 0.163). Conclusion This study reported that FIB-4 displayed the highest correlation with liver stiffness measurements obtained through transient elastography in detecting liver fibrosis. APRI and AARPRI reported a moderate correlation, while AAR and BARD were less consistent in their performance. These findings suggest that FIB-4, APRI, and AARPRI are effective non-invasive tools for assessing liver fibrosis in a limited resource setting where transient elastography is not available.