Diagnostic accuracy of two-dimensional shear wave elastography for the non-invasive staging of hepatic fibrosis in chronic hepatitis B: a cohort study with internal validation.

Abstract

To determine the accuracy of two-dimensional shear wave elastography (2D-SWE) for noninvasive staging of hepatic fibrosis in chronic hepatitis B (CHB). Patients with CHB infection who underwent liver biopsy were consecutively included. Receiver-operating characteristic (ROC) curves were constructed to assess the overall accuracy and identify optimal cutoff values. Three hundred three patients were analysed. The diagnostic performance characteristics were determined for the first 202 patients (the index cohort) and were validated on the next 101 patients (validation cohort). The areas under the ROC curves for significant fibrosis, severe fibrosis and cirrhosis were all greater than 0.90 and did not differ significantly between the index and validation cohorts. Using the cutoff values generated from the index cohort, the validation cohort 2D-SWE had negative predictive values of 82.6 % (95 % confidence interval [CI]: 68.4 % - 92.3 %) for significant fibrosis, 95.1 % (95 % CI: 86.3 % - 99.0 %) for severe fibrosis and 97.4 % (95 % CI: 90.8 % - 99.7 %) for cirrhosis. The positive predictive values were 83.6 % (95 % CI: 71.2 % - 92.2 %), 65.0 % (95 % CI: 48.1 - 79.5 %) and 60.0 % (95 % CI: 38.7 % - 78.9 %), respectively. The 2D-SWE showed good diagnostic accuracy in staging liver fibrosis in patients with CHB infection and assisted in excluding liver fibrosis and cirrhosis. • Two-dimensional shear wave elastography showed good diagnostic accuracy in assessing liver fibrosis. • Diagnostic performance did not differ significantly between the index and validation cohorts. • Two-dimensional shear wave elastography assisted in excluding liver fibrosis and cirrhosis.