Abstract
BACKGROUND: Differentiating biliary atresia (BA) from non-BA is paramount, as their management and outcome differ. So far, BA is diagnosed by intraoperative cholangiogram (IOC) or liver biopsy. However, literature is sparse on the utility of transient elastography (TE) in neonatal cholestasis. We aimed to study the usefulness and accuracy of TE in differentiating BA from other causes of neonatal cholestasis (NCS). METHODS: A prospective observational study was conducted from June 2021 to February 2024; consecutive cases of NCS were recruited. Patients were grouped as BA and non-BA. All patients underwent clinical, laboratory, and radiological evaluation, along with TE. Receiver operating characteristic curve (ROC) analysis was performed to derive an LSM cutoff for diagnosing BA. The liver stiffness measurement (LSM) was correlated with histopathological fibrosis. RESULTS: We enrolled 135 patients, BA (n = 68) and non-BA (n = 67). The median LSM was higher in BA, 19.4 (12.9-38.75) kPa vs. 12.0 (8.4-17.9) kPa (P=<0.001). LSM positively correlated with the METAVIR fibrosis score (Spearman's correlation coefficient, r = 0.590, P < 0.001). We derived an LSM cutoff of 13.1 kPa to differentiate BA from non-BA (sensitivity 75%, specificity 61.2%, AUC 0.736). By adding ultrasound (USG) finding of a small gallbladder (<19 mm) to our LSM cutoff (13.1 kPa), the sensitivity improved to 97.0%, but specificity remained low (52.2%). CONCLUSION: LSM correlates well with histopathological fibrosis in NCS. An LSM cut off of 13.1 kPa can be used to differentiate BA from non-BA with a sensitivity of 75%. In resource-constraint setting where liver biopsy and IOC may not be available, TE in combination with USG can be a good screening tool for BA (sensitivity 97%).